Relation between the anatomical genitalphenotype and cystic fibrosistransmembrane conductance regulatorgene mutations in the absence of the vasdeferens

Francois Robert, M.D.,a Faıza Bey-Omar, M.D.,b Jacques Rollet, M.D.,a

Jean-Francois Lapray, M.D.,c and Yves Morel, M.D., Ph.D.b

Institut Rhonalpin pour la Reproduction Humaine, Lyon-Bron, and Laboratoire de Biochimie Endocrinienne etMoleculaire, Hopital Debrousse, Lyon, France

Objective: To study the correlation between genital phenotype and cystic fibrosis genotype in men lackingat least one vas deferens.

Design: Prospective study.

Setting: Institut Rhonalpin pour la Reproduction Humaine, Lyon-Bron, France.

Patient(s): Forty-seven infertile men lacking at least one vas deferens.

Intervention(s): All patients were screened for the 13 most common CFTR gene mutations and for the5-thymidine variant of intron 8. Renal, scrotal, and transrectal ultrasonography were systematically performed.

Main Outcome Measure(s): Epididymal and seminal vesicular abnormalities and testicular volume werecompared among men with two, one, or no CFTR gene mutation, with or without the 5T allele.

Results: Seminal vesicles and the symmetry of epididymal and vesicular abnormalities did not differ betweenpatients with and those without the CFTR gene mutation. Epididymal abnormalities were more frequent inmen without the mutation. Testicular volumes were significantly lower in men without the mutation and thosewith the 5T allele only.

Conclusion: Men with the CFTR mutation, the 5T allele only, and those without CFTR mutation have fewdifferences in genital phenotype. Low testicular volume is observed in men without the CFTR mutation andthose with the 5T allele only. (Fertil Steril� 2002;77:889–96. ©2002 by American Society for ReproductiveMedicine.)

Key Words: Male infertility, azoospermia, vas deferens, CFTR gene

Congenital bilateral absence of the vas de-ferens (CBAVD) is a well-known cause ofmale sterility; it is found in 1%–2% of meninvestigated for infertility and approximately10% of men with azoospermia (1). The etiol-ogy of CBAVD remains unknown. Most malepatients with cystic fibrosis have CBAVD, andit has been suggested that CBAVD representsan incomplete form of cystic fibrosis. The cys-tic fibrosis transmembrane conductance regu-lator (CFTR) gene was identified in 1989, andDumur et al. (2) established the link betweenabsence of the vas deferens and mutations inthe CFTR gene in 1990. These investigatorsreported that 7 of 17 men with isolated

CBAVD had the �F508 mutation in the CTFRgene.

Cystic fibrosis is a common severe autoso-mal recessive disease that affects 1 in 2,500people among white persons. The risk for het-erozygosity in the general population is 1 in 25persons. More than 900 mutations of the CFTRgene have been identified (3), which encodesfor a protein containing 1,480 amino acids.Cystic fibrosis is characterized by abnormalflux of chloride in the apical membrane ofepithelial cells, leading to a diverse clinicalpresentation that includes pancreatic insuffi-ciency, pulmonary disease, meconium ileus,elevated sweat levels of chloride, and absence

Received June 18, 2001;revised and acceptedOctober 29, 2001.Reprint requests: FrancoisRobert, M.D., InstitutRhonalpin pour laReproduction Humaine, 1rue Laborde, 69500 Lyon-Bron, France (FAX: 33-4-78-77-58-02; E-mail:fran.robert@wanadoo.fr).a Institut Rhonalpin pour laReproduction Humaine.b Laboratoire de BiochimieEndocrinienne etMoleculaire, INSERMU329, Hopital Debrousse.c Cabinet de Radiologie etd’Echographie, Lyon,France.

MALE FACTORFERTILITY AND STERILITY�VOL. 77, NO. 5, MAY 2002Copyright ©2002 American Society for Reproductive MedicinePublished by Elsevier Science Inc.Printed on acid-free paper in U.S.A.

0015-0282/02/$22.00PII S0015-0282(02)02954-0

889

of the vas deferens. This last characteristic is observed inmore than 95% of men with cystic fibrosis.

Clinical symptoms of CBAVD are bilateral nonpalpablevas deferens, absence of the distal part of or the wholeepididymis, and hypotrophy or absence of the seminal ves-icles, leading to azoospermia with low semen plasma volumeand acidic pH.

Mutation analysis of the CFTR gene in CBAVD hasshown that about 50% to 80% of men with CBAVD carry atleast one mutation (4–8). These variations are explained bythe different methods used to screen the CFTR gene muta-tions: Either the most common ones (about 10 to 30) or allexons are analyzed.

The inability to identify a second mutation in mostpatients with CBAVD, even after analysis of the entirecoding region, suggested that mutations may be located inthe noncoding regions of the gene. Such an allele termedthe 5T allele has been identified in intron 8 (IVS8-5T).This sequence contains five, seven, or nine thymines. Thefrequency of skipping exon 9 has been found to be in-versely proportional to the number of thymines. The 5Tallele results in reduced levels of normal CFTR messenger(m)RNA because of deletion of exon 9 and is considereda mild variant. Patients who are homozygous for the 5Tallele have nonfunctional CFTR mRNA that accounts for92% of the total mRNA (9).

Genotype–phenotype studies in cystic fibrosis have iden-tified two categories of alleles on the basis of pancreaticexocrine function: mild or severe. More than 80% of patientswith cystic fibrosis carry two severe mutations, whereasabout 60% of men with CBAVD have a severe and a mildmutation and 10% have two mild mutations (10–12). How-ever, as we have previously reported (13), absence of the vasdeferens follows different patterns: It may be unilateral orbilateral and with or without unilateral renal agenesis, sem-inal vesicles, and CFTR gene mutation.

Mutations of the CFTR gene are reported in only about17% of men with congenital unilateral absence of a vasdeferens (CUAVD). Unilateral agenesis in about 50% ofmen with CUAVD vs. about 10% of men with CBAVD. Itwas thought that when renal agenesis is present, no mutationon the CFTR gene was usually observed (14); however,Casals et al. (8) detected one CFTR gene mutation in onethird (5 of 16) of patients with congenital absence of the vasdeferens (CAVD) and renal agenesis (3 with CUAVD and 2with CBAVD).

Because the severity of cystic fibrosis is to some extentrelated to CFTR genotype, we tested this correlation betweenCAVD and CFTR genotype. We evaluated the role of dif-ferent CFTR gene mutations in morphologic genital pheno-type of 47 infertile men with absence of the vas deferens.

MATERIALS AND METHODS

PatientsForty-seven men being treated for infertility for at least

for 1.5 years were evaluated at the Institut Rhonalpin pour laReproduction Humaine, Lyon, France. After a thorough his-tory was obtained, all patients had a complete physicalexamination by the same physician, with particular attentionto the testes, vasa deferentia, epididymides, and prostategland (evaluated by digital rectal examination). Testicularvolume was evaluated by orchidometry. All patients lackedat least one vas deferens. Each patient gave informed consentfor semen analysis, ultrasonographic examination, andCFTR genotype analysis. The study was approved by theinstitute’s review board.

Semen analysis and a serum hormonal profile, includingradioimmunologic assay (Bio-Merieux, Lyon, France) ofFSH, LH, prolactin, and testosterone, were also performed.

Imaging of the Reproductive Tract andKidneys

Ultrasonographic examination was performed in all pa-tients by the same radiologist using two Bruel Kjaer(Naerum, Denmark) ultrasonographs (1849 and 3535). Thekidneys and vesica were examined percutaneously by usinga 3.5-MHz transducer. The testes and epidimydes were ex-amined by using a 7.5-MHz transducer. Testicular volumewas evaluated by measurement of width (W) and length (L)and was calculated as an ellipsoidal volume by using thefollowing equation: 0.52 � L � W2. Transrectal ultrasonog-raphy was performed by using a 7-MHz biplanar transducerinserted into the rectum. Images were obtained in the trans-axial and sagittal planes. They detailed the ampulla of thevas deferens (absent), seminal vesicles (absent, hypotrophic,or dilatated if transverse width was � 1 cm), ejaculatoryducts (dilatation or calcification), and prostate gland.

Genetic Screening for CFTR Gene MutationsAs previously reported, the 13 most frequent CFTR gene

mutations and the thymidine tract length of the splice accep-tor site in intron 8 (IVS8-polyT) were tested in a whiteFrench sample. In 950 unrelated chromosomes, the inci-dences of �F508, G542X and N1303K mutations were 69%,4.6% and 2.7% respectively (15, 16; unpublished data). Theincidence of all other mutations was less than 1%. Thus,results of this screening represents about 82% of cysticfibrosis chromosomes. Genomic DNA was extracted from 25mL of whole blood preserved in ethylenediamine tetraaceticacid by using a modified phenol–chloroform extraction tech-nique described by Morel et al. (17). By using intron oligo-nucleotides that bracketed exons according to the CysticFibrosis Genetic Analysis Consortium (3), exons 3, 4, 7, 10,11, 14b, 17b, 20, and 21 were amplified with polymerasechain reaction (PCR) as described elsewhere (18). Methodsused to detect these mutations were [1] heteroduplex forma-tion followed by polyacrylamide gel electrophoresis (�F508,

890 Robert et al. Absence of the vas deferens Vol. 77, No. 5, May 2002

�I507, 508C, and 1612delTT in exon 10 and 2183AA3G,2184delA, and 2347delG in exon 13), [2] digestion withappropriate restriction enzymes, that is, MnlI for W1282X(exon 20) and SspI for 2789�5G3A (exon 14b), and [3]PCR with the modified primers MvaI for G542X (exon 11)and N1303K (exon 21), AvaII for 1717-1G3A (exon 11),and HaII for R117H (exon 4) to create a new restriction site.The polypyramidine sequence length of the splice acceptorsite in intron 8 was analyzed by PCR amplification of exon9, followed by a nested PCR in patients with CAVD, asdescribed elsewhere (4). The final PCR products were di-gested by XmnI and visualized by using nondenaturing poly-acrylamide gel electrophoresis (8%). CFTR gene mutationsfrom 950 unrelated cystic fibrosis chromosomes were usedfor comparison with the CFTR genotype of patients withCAVD.

Statistical AnalysisSeven subgroups to compare reproductive tract abnormal-

ities: two CFTR gene mutations, one CFTR gene mutationand 5T allele, one CFTR gene mutation, two 5T alleles, 5Talone, no mutation, and absence of the vas deferens withunilateral renal agenesis (no mutation). Differences betweenpercentages were tested by using the �2 test. Means (�SD)were compared by using the Student t-test. P�.05 wasconsidered statistically significant.

RESULTS

Forty-seven infertile men lacked at least one vas deferens,as determined by scrotal palpation. Absence of the vasdeferens was confirmed by absence of the distal portion ofthe vas deferens (ampulla) on transrectal ultrasonography.Of the 47 patients, 40 had isolated CBAVD, 2 of whom hadunilateral renal agenesis. Seven patients had CUAVD, ofwhom 1 had unilateral renal agenesis. Table 1 shows clini-cal, ultrasonographic, and genetic characteristics of the pa-tients.

CFTR MutationsOne or two CFTR gene mutations, including the 5T allele,

were detected in 30 of 47 patients (64%). The two mostfrequent mutations were the 5T allele (19 of 30 patients[63%]) and �F508 (14 of 30 patients [47%]). Mutationswere not detected in 17 patients, including the 3 patients withunilateral renal agenesis. Three patients had two CFTR genemutations: two had the severe mutation �F508 associatedwith the mild mutation R117H, and one had two mild mu-tations (R117H and R347H).

Three patients were found to be compound heterozygotesafter CFTR gene analysis of parents or siblings. When wecompared the 30 men with CAVD who had CFTR mutation(60 chromosomes) and 950 unrelated CF chromosomes,allele frequencies differed between cystic fibrosis andCAVD chromosomes (Table 2). Patients with cystic fibrosis

carry severe mutations, whereas men with CAVD carry oneor two mild mutations, including the 5T allele.

Reproductive Tract Phenotype According toCFTR Gene Mutation

Three patients with two CFTR gene mutations lacked atleast one seminal vesicle, and the other one, if present, wasdilated. Eight patients with �F508 and the 5T allele had anabsent, hypotrophic or dilated seminal vesicle. Eight patientshad only one CFTR mutation and no 5T allele; they lackedat least one seminal vesicle, and abnormalities were asym-metric in 50%. One patient had two 5T alleles; one seminalvesicle was absent, and the other one was dilated. Bothseminal vesicles were absent in 5 of the 7 patients with 5Tallele only (70%). Of the 13 men in whom no mutation wasdetected, 2 had unilateral renal agenesis and the seminalvesicles were absent or hypotrophic in 77%. Of the 7 menwith unilateral absence of the vas deferens, 3 had the 5Tallele and no other CFTR gene mutation. Two of these 7 menhad a controlateral testicular abnormality (1 had hypotrophyand 1 had cryptorchidism).

Patients 44, 45, and 46, who had no mutation and no renalagenesis, had an FSH level above the normal range (11.6 U/L,12.6 U/L, and 18.5 U/L, respectively) (normal, �7 U/L).

Correlation Analysis Between Phenotype andGenotype

The 47 patients were classified into seven subgroupsaccording to genotype in order to compare morphologicgenital findings (Tables 3 and 4). Subgroups did not differ inseminal vesicle status and symmetry of vesicular and epi-didymal abnormalities. Epididymal abnormalities were sig-nificantly more frequent in the group without mutation thanin the group with CFTR gene mutation, including the 5Tallele (73% vs. 41%, respectively; P�.01, �2 test). Testicularvolume was smaller in men with CBAVD and no mutationand in men with CUAVD than in men with CFTR mutation(Fig. 1). The difference between testicular volume in patientswith at least one CFTR gene mutation and those withoutmutation or with the 5T allele only differed significantly(15.1 � 4.5 mL vs. 10.7 � 4.1 mL, respectively; P�.001,Student t-test). This difference remained significant after thepatient with atrophic testicles (number 26) and the patientwith cryptorchidism (number 44) were excluded.

Scrotal ultrasonography found dilatation of the rete testisin only one patient and intratesticular calcification in nopatient.

DISCUSSION

We found that 30 patients with CAVD and a CFTRmutation had the same seminal vesicle phenotype as 17patients without CFTR mutation. Epididymal abnormalitieswere even more frequent in men without mutation. Thesedata are similar to previous findings in patients with CAVD.

FERTILITY & STERILITY� 891

Jarvi et al. (19) reported that 39 of 46 (85%) men withCBAVD had at least one CFTR gene mutation or the 5Tallele and 13 (28%) had two CFTR gene mutations.

Among 33 men studied with scrotal ultrasonography, thefrequency of intrascrotal abnormalities (testicular calcification,testicular atrophy, cystic dilatation of rete testis, or varicocele)

did not differ among subgroups of men with or without CFTRgene mutations. Transrectal ultrasonography in 25 men re-vealed ampullae of the vas deferens in 26%, and 80% hadatrophic seminal vesicles. Correlation between CFTR gene mu-tation and results of transrectal ultrasonography showed that50% (2 of 4) of men without CFTR gene mutations have normal

T A B L E 1

Genetic and ultrasonographic findings in 47 patients with absence of the vas deferens.

Patient Mutations Intron 8 Phenotype Seminal vesicle Epididymis Testicular volume (right/left) (mL) Comments

1 �F508/R117H 7T/9T BAVD A/D C/C 20/202 �F508/R117H 7T/9T BAVD A/A E/C 15/123 R117H/R347H 7T/9T BAVD A/D E/E 15/15 Compound heterozygote4 �F508/o 5T/7T BAVD A/h C/C 10/13 Compound heterozygote5 �F508/o 5T/9T BAVD A/A E/E 16/17 Compound heterozygote6 �F508/o 5T/9T BAVD A/D E/E 15/157 �F508/o 5T/9T BAVD A/A E/E 11/98 �F508/o 5T/9T BAVD h/h E/E 20/209 �F508/o 5T/9T BAVD A/A C/C 17/9

10 �F508/o 5T/9T BAVD D/A E/E 15/1211 �F508/o 5T/9T BAVD A/D E/E 10/1212 �F508/o 7T/9T BAVD A/N C/E 20/2013 �F508/o 7T/7T BAVD A/A E/E 20/1814 �F508/o 7T/7T BAVD h/h E/E 10/915 �F508/o 9T/9T BAVD A/D C/C 23/2616 1612delTT/o 7T/9T BAVD A/D C/C 8/717 D1270N/o 7T/7T BAVD A/A E/E 12/1318 R117H/o 7T/7T BAVD A/D E/E 20/1819 2789�5GBA/o 7T/7T BAVD A/A C/C 18/1320 o/o 5T/5T BAVD D/A E/E 8/1021 o/o 5T/9T BAVD A/A C/E 20/1822 o/o 5T/7T BAVD A/A E/E 20/2023 o/o 5T/9T BAVD A/A E/E 9/1024 o/o 5T/7T BAVD A/D E/E 10/1025 o/o 5T/7T BAVD A/A E/E 9/1026 o/o 5T/9T BAVD A/D E/A 12/0 Left testicular atrophy27 o/o 5T/7T BAVD A/A C/E 9/828 o/o 7T/7T BAVD A/A E/E 15/1529 o/o 7T/7T BAVD A/h E/E 11/1130 o/o 9T/9T BAVD h/h C/C 7/831 o/o 7T/7T BAVD h/A E/E 12/1232 o/o 7T/7T BAVD A/A E/C 10/1033 o/o 7T/9T BAVD A/A E/E 11/1234 o/o 7T/7T BAVD h/A C/C 9/935 o/o 7T/7T BAVD A/D C/A 10/11 Bilateral dilatation of rete testis36 o/o 7T/7T BAVD A/D C/C 12/1237 o/o 7T/9T BAVD A/A C/E 15/1538 o/o 7T/7T BAVD D/A C/C 12/1539 o/o 7T/7T BAVD�URA A/A C/C 12/1240 o/o 7T/7T BAVD�URA D/A A/C 0/20 Right testis absent41 o/o 5T/9T UAVD D/A E/E 13/1342 o/o 5T/7T UAVD D/A E/C 8/10 Right testicular hypotrophy43 o/o 5T/7T UAVD h/N E/E 7/844 o/o 7T/7T UAVD A/D E/A 12/0 Left cryptorchidism45 o/o 7T/7T UAVD D/A E/E 4/5 Bilateral testicular hypotrophy46 o/o 7T/7T UAVD h/h E/E 8/847 o/o 7T/7T UAVD�URA D/A E/E 15/15

Note: A � absent; D � dilated; C � caput only; E � whole; h � hypotrophic; N � normal; o � no detected mutation; BAVD � bilateral absence of thevas deferens; UAVD � unilateral absence of the vas deferens; URA � unilateral renal agenesis.

Robert. Absence of the vas deferens. Fertil Steril 2002.

892 Robert et al. Absence of the vas deferens Vol. 77, No. 5, May 2002

seminal vesicles and vas deferens and more than 80% of 21men with at least one CFTR gene mutation have atrophicseminal vesicles; these findings did not differ significantlyamong the subgroups with the 5T allele only, one mutation, ortwo mutations.

De la Taille et al. (20) described 41 patients with CBAVDconfirmed by surgical exploration. Analysis for the 5T alleleand scrotal ultrasonography were not done. No CFTR gene

mutation was detected in 8 patients with renal abnormalities.Seminal vesicle abnormalities were not related to the pres-ence or absence of mutations (8 of 23, patients vs. 11 of 18patients, respectively) or compound heterozygosity (8 of 13patients vs. 3 of 10 patients). Daudin et al. (21) found nocorrelation between epididymal and seminal vesicule mor-phology and CFTR gene mutations in 32 patients screenedfor 10 mutations, exon 4, and the 5T allele.

T A B L E 2

Allele frequencies of six, CFTR gene mutations in unrelated cystic fibrosis chromosomes, CBAVD chromosomes, andCUAVD chromosomes.

MutationCystic fibrosis chromosome

(n � 950) (%)CBAVD chromosome

(n � 54) (%)CUAVD chromosome

(n � 6) (%)General

population (%)a

�F 508b 69 25.9 0 1.4G542Xb 4.6 0 0 —N1303Kb 2.7 0 0 —1717-1G�Ab 0.9 0 0 —R117H 0.3 7.4 0 —5T — 31.5 50 5.2

Note: CBAVD � congenital bilateral absence of the vas deferens; CUAVD � congenital unilateral absence of the vas deferens.a Data (4).b These mutations are considered severe.

Robert. Absence of the vas deferens. Fertil Steril 2002.

T A B L E 3

Ultrasonographic findings in seminal vesicles, epididymides, and testicular volume and CFTR genotype in 40 patients withBAVD.

BAVDTwo mutations

(n � 6)CFTR/5T(n � 16)

One mutation(n � 16)

5T/5T(n � 2)

5T/o(n � 14)

o/o(n � 22)

URA(n � 4)

Seminal vesiclesOne normal 1One dilated 2 3 3 1 2 3Both dilated 1One hypotrophic 1 3 1Both hypotrophic 1 1One absent 2 4 4 1 2 6 1Both absent 1 3 3 5 4 1At least one

absent (%)100 87.5 87.5 100 100 91 100

EpididymidesOne cauda 1 1 3 2Both cauda 1 6 4 1 4 3One caput 1 1 2 3 1Both caputs 1 2 3 5 1Absence 1 1 1At least one 67 25a 50 0 43 73b 100

abnormal (%) 41a

Testicular volume(mL)

16.2 � 2.9 13.8 � 3.5 15.9 � 5.5 9 � 1 11.8 � 5.5 11.5 � 2.3 11.5 � 3.5

Note: Data are the number of patients, unless otherwise stated. CBAVD � congenital bilateral absence of vas deferens; o � no mutation detected; URA �unilateral renal agenesis.b P�.01 vs. subgroup a (�2 test).

Robert. Absence of the vas deferens. Fertil Steril 2002.

FERTILITY & STERILITY� 893

Testicular volume is small in about 5% to 10% of patientswith CBAVD (1, 19, 21), but a correlation between testissize and CFTR mutations was never established. We found asignificant correlation between testicular volume and having

at least one CFTR gene mutation or having no mutation orthe 5T allele only; testicular volume was smallest in the lastgroup. Moreover, the three patients with CUAVD and nomutation had testicular hypotrophy (one had cryptorchidism)and serum FSH levels slightly higher than the normal range,suggesting that a testicular factor (genetic or environmental)plays a role in determining the phenotype. Further studiesand standardization of testicular volume are needed to con-firm these observations.

Pathogenetic Implications“Agenesis” is not the most appropriate term to include in

the acronym “CBAVD.” Several factors suggest that invo-lution or atresia of the vas deferens, sometimes including theseminal vesicles and the epididymides, rather than agenesisoccurs in CBAVD.

First, bilateral wolffian duct agenesis would result indeath by the 28th day of gestation, due to lack of formationof the kidneys from the ureteral buds.

Second, Gaillard et al. (22) examined two male fetuseswith cystic fibrosis after abortion at 12 and 18 weeks respec-tively. The fetuses had the �F508/�F508 and �F508/G542Xgenetic mutations. The vas deferens of these fetuses withcystic fibrosis showed normal development, and secretionswere present in the lumen of the vas deferens. No obstruc-tion, epithelial necrosis, focal dilatation, or fibrous stenosiswas detected at any age. These observations and the highproportion of normal ducts reported in prepubertal maleswith cystic fibrosis suggest a mechanism of luminal obstruc-tion that results in duct atrophy or atresia (23–25).

Third, during organogenesis, the development of wolffianducts precedes and is required for the development of mul-lerian ducts. However, in female patients with cystic fibrosis,no genital tract abnormalities have been reported (26).

Finally, in the reproductive tissues of male fetuses, sig-nificant expression of CFTR mRNA is detected in the epi-thelium of the epididymis at all stages of development and inthe vas deferens. CFTR expression is most prominent in thehead of the epididymis, and no expression is found in thetestis, the seminal vesicles, and the prostate (27–29). Thelow expression of CFTR in the epididymis may indicate thatits development is more sensitive to CFTR function than areother organs of the fetus. The luminal fluid of the epididymiscontains a high protein load (albumin, �2-macroglobulin,transferrin, and androgen-binding protein) and has a lowflow rate; the absence of CFTR function may lead to ob-struction and obliteration, a similar mechanism described inother ducts in cystic fibrosis. Moreover, the epididymis andthe vas deferens form a tortuous ductal system that is 7meters long but has a diameter less than 0.5 mm. Theabsence or dysfunction of CFTR would make these organsvery vulnerable to luminal concentration defects, especiallyin the distal portion of the vas deferens, where CFTR ex-pression level is low.

T A B L E 4

Ultrasonographic findings on seminal vesicles,epididymides, and testicular volume and CFTR genotype inseven patients with CUAVD.

CUAVD 5T/� (n � 6) o/o (n � 6) URA (n � 2)

Seminal vesiclesOne normal 1One dilated 2 2 1One hypotrophic 1Both hypotrophic 1One absent (%) 33 (%) 33 (%) 50 (%)

EpididymidesBoth cauda 2 2 1One cauda 1 1One caput 1absent 1At least one abnormal (%) 33 33 0

Testicular volume (mL) 9.8 � 4.9 6.2 � 3.8 15 � 0

Note: Data are the number of patients, unless otherwise indicated.CUAVD � congenital unilateral absence of the vas deferens; o � nomutation detected; URA � unilateral renal agenesis.

Robert. Absence of the vas deferens. Fertil Steril 2002.

F I G U R E 1

Testicular volume according to CFTR gene status in 44 menwith absence of the vas deferens and no renal agenesis. Pvalue is for comparison with the CFTR gene mutation group.The horizontal line represents the mean. n � 88 testis.

Robert. Absence of the vas deferens. Fertil Steril 2002.

894 Robert et al. Absence of the vas deferens Vol. 77, No. 5, May 2002

Thus, in CBAVD, the broad spectrum of reproductivetract abnormalities seems to be determined by a randominvolution of the vas deferens that does not always involvethe seminal vesicles and the epididymides. This processdepends in part on CFTR mutation in most cases. Otherfactors that determine or allow obstruction, then involution,of the vas deferens remain unknown. Only 1% to 2% of menwho are heterozygous for one CFTR gene mutation lack avas deferens, indicating that other genetic (e.g., a mild CFTRgene mutation in a compound heterozygote) or environmen-tal factors (e.g., infectious genital disease) are required todevelop such abnormalities. Obstructive factors may there-fore explain the absence of the vas deferens in men withouta CFTR gene mutation, and testicular factors may explain thelow testicular volume reported in these men. This hypothesisis represented in Figure 2, which shows the respective role ofeach pathogenic factor.

Patients with CAVD and renal agenesis have the samereproductive tract abnormalities as do those with isolatedCAVD, but the genetic basis that leads to abnormal devel-opment of the entire mesonephric duct before week 7 ofgestation remains unidentified (30).

In conclusion, we found that reproductive tract abnormal-ities are independent of the subtypes of CFTR genotype inmen with absence of the vas deferens and CFTR genemutations, but men with the 5T allele only have smallertesticular volume. Men without CFTR gene mutations hadmore epididymal abnormalities and smaller testicular vol-ume. These different patterns may share some of the samepathogenic factors. Further studies using common diagnosticcriteria are required to confirm the observations.

References1. Jequier AM, Ansell ID, Bullimore NJ. Congenital absence of the vasa

deferentia presenting with infertility. J Androl 1985;6:15–9.

2. Dumur V, Gervais R, Rigot JM, Lafitte JJ, Manouvrier S, Biserte J, etal. Abnormal distribution of CF �F 508 allele in azoospermic men withcongenital aplasia of epididymis and vas deferens. Lancet 1990;336:512.

3. Cystic fibrosis mutation data base. http://www.genet.sickkids.on.ca/cftr/.

4. Chillon M, Casals T, Mercier B, Bassas L, Lissens W, Silber S, et al.Mutations in the cystic fibrosis gene in patients with congenital absenceof the vas deferens. N Engl J Med 1995;332:1475–80.

5. Oates RD, Amos JA. The genetic basis of congenital bilateral absenceof the vas deferens and cystic fibrosis. J Androl 1994;15:1–8.

6. Augarten A, Yahav Y, Kerem BS, Halle D, Laufer J, Szeinberg A, et al.Congenital bilateral absence of vas deferens in the absence of cysticfibrosis. Lancet 1994;344:1473–4.

7. Dork T, Dworniczak B, Aulehla-Scholz C, Wieczorek D, Bohm I,Mayerova A, et al. Distinct spectrum of CFTR gene mutations incongenital absence of the vas deferens. Hum Genet 1997:100;365–77.

8. Casals T, Bassas L, Egozcue S, Ramos MD, Gimenez J, Segura A, etal. Heterogeneity for mutations in the CFTR gene and clinical correla-tions in patients with congenital absence of the vas deferens. HumReprod 2000:15;1476–83.

9. Chu CS, Trapnell BC, Curristin S, Cutting GR, Crystal RG. Geneticbasis of variable exon 9 skipping in cystic fibrosis transmembraneconductance regulator mRNA. Nat Genet 1993:3;151–6.

10. The Cystic Fibrosis Genotype-Phenotype Consortium. Correlation be-tween genotype and phenotype in patients with cystic fibrosis. N EnglJ Med 1993;329:1308–13.

11. Gan KH, Veeze HJ, van den Ouweland AMW, Halley DJJ, Scheffer H,van der Hout A, et al. A cystic fibrosis associated with mild lungdisease. N Engl J Med 1995;333:95–9.

12. Claustres M, Guittard C, Bozon D, Chevalier F, Verlingue C, Ferec C,et al. Spectrum of CFTR mutations in the cystic fibrosis and in con-genital absence of the vas deferens in France. Hum Mutat 2000;16:143–56.

13. Robert F, Rollet J, Lapray JF, Bey-Omar F, Durieu I, Morel Y. Lesabsences de canaux deferents dans l’infertilite masculine: tentative declassification a partir de 39 cas. Press Med 1999;28:116–21.

14. Augarten A, Yahav Y, Kerem BS, Halle D, Laufer J, Szeinberg A, et al.Congenital bilateral absence of vas deferens in the absence of cysticfibrosis. Lancet 1994;344;1473–4.

15. Population variation of common cystic fibrosis mutations. The CysticFibrosis Genetic Consortium. Hum Mutat 1994;4:167–77.

16. Estivill X, Bancells C, Ramos C. Geographic distribution and re-gional origin of 272 cystic fibrosis mutations in European populations.The Biomed CF Mutation Analysis Consortium. Hum Mut 1997;10:135–54.

17. Morel Y, Andre J, Uring-Lambert B, Hanptmann G, Betuel H, Tosi M,et al. Rearrangements and point mutations of P450C21 genes aredistinguished by five restriction endonuclease haplotypes identified bya new probing strategy in 57 families with congenital adrenal hyper-plasia. J Clin Invest 1989;83:527–36.

18. Durieu I, Bey-Omar F, Rollet J, Calemard L, Boggio D, Lejeune H, etal. Diagnostic criteria for cystic fibrosis in men with congenital absenceof the vas deferens. Medicine (Baltimore) 1995;74:42–7.

19. Jarvi K, McCallum S, Zielenski J, Durie P, Tullis E, Wilchanski M, etal. Heterogeneity of reproductive tract abnormalities in men with ab-sence of the vas deferens: role of cystic fibrosis transmembrane con-ductance regulator gene mutations. Fertil Steril 1998;70:724–8.

20. De la Taille A, Rigot JM, Mahe P, Vankemmel O, Gervais R, DumurV, et al. Correlation between genito-urinary anomalies, semen analysisand CFTR genotype in patients with congenital bilateral absence of thevas deferens. Br J Urol 1998;81:614–9.

21. Daudin M, Bieth E, Bujan L, Massat G, Pontonnier F, Mieusset R.Congenital bilateral absence of the vas deferens: clinical characteristics,biological parameters, cystic fibrosis transmembrane conductance reg-ulator gene mutations and implications for genetic counseling. FertilSteril 2000;74:1164–74.

22. Gaillard DA, Carre-Pigeon F, Lallemand A. Normal vas deferens infetuses with cystic fibrosis. J Urol 1997;158:1549–52.

23. Landing BH, Wells TR, Wang CI. Abnormality of the epididymis andvas deferens in cystic fibrosis. Arch Pathol 1969;88:569–80.

24. Oppenheimer EH, Esterly JR. Observations on cystic fibrosis of thepancreas. V. Developmental changes in the male genital system. J Pe-dict 1969;75:806–11.

25. Valman HB, France NE. The vas deferens in cystic fibrosis. Lancet1969;2:566–7.

26. Oates RD, Amos JA. The genetic basis of congenital bilateral absenceof the vas deferens and cystic fibrosis. J Androl 1994;15:1–8.

27. Tizzano EF, Silver M, Chitayat D, Benichou JC, Buchwald M. Differ-

F I G U R E 2

Schematic representation of the roles of CFTR gene muta-tions and hypothetical factors in the pathogenesis of ab-sence of the vas deferens.

Robert. Absence of the vas deferens. Fertil Steril 2002.

FERTILITY & STERILITY� 895

ential cellular expression of CFTR in human reproductive tissues: cluesfor the infertility in patients with CF. Am J Pathol 1994;144:906–14.

28. Tizzano EF, Chitayat D, Buchwald M. Cell-specific localization ofCFTR mRNA shows developmentally regulated expression in humanfetal tissues. Hum Mol Genet 1993;2:219–22.

29. Trezise AEO, Chambers JA, Wardle CJ, Gould S, Harris A. Expression

of the cystic fibrosis gene in human fetal tissues. Hum Mol Genet1993;2:213–18.

30. McCallum TJ, Milunsky JM, Munarriz R, Carson R, Sadeghi-Nejad H,Oates RD. Unilateral renal agenesis associated with congenital bilateralabsence of the vas deferensph: phenotypic findings and genetic consid-erations. Hum Reprod 2001;16:282–8.

896 Robert et al. Absence of the vas deferens Vol. 77, No. 5, May 2002