`
`(
`
`I
`' I
`I
`
`{
`
`Adult and
`Pediatric
`Urology
`
`THIRD EDITION
`
`EDITORS
`Jny Y. Gillenwater, :\l.D.
`Hovey S. Dabney Professor
`Department of Urology
`University of Virginia l-leruth Sciences Center
`Charlottcsv;lle, Virginia
`Joh n T. Grayhack, M.D.
`Professor
`Dcpa1·trnent of Urology
`Northwestern University Mcd irol School
`Chicago, 1llinois
`
`Stuart S. Howards, M.D.
`Professor
`Department of Urology
`University of Virginia Health Sciences Center
`Charlottesville, Virginia
`
`John W. Duckett, M.D.
`P1·ofossor of Urology
`University of Pennsylvnnin School of Medicine;
`Director, Pediatric Urology
`Children's Hospital of Philadelphia
`Philadelphia, Peonsylvanio
`
`With 16<12 illustmtions
`
`~·A Mosby
`
`Si Lou.s 8.1!bmoro Bostoo c,,1,b.ld Chicago MapleS Htw Yan Ptilladets:hia Porlbnd
`London MaJn!l Max!co Cd., S l'liPPolt S·r.lney
`lokyo Tororao \V1tsbit~en
`
`West-Ward Exhibit 1043
`Jennings & Linehan 1996
`Page 001
`
`
`
`.....,. J\ Ti1nes ;\1irror
`M Cu1np~111y
`
`Publisher: Anne S. Patterson
`Editor: Susie Baxter
`DevelopmcnUil EditOI': 1\nne Gunter
`Projecl Maun.ger: Peggy fo'agen
`
`THIBD EDITTON
`Copyright o 1996 by Mosby- Year Book, Inc.
`
`Previous editions copyrighted 1987, 1991
`
`All rights reserved. No P'trt of this publication may be reproduced,
`stored in a retrieval system, or tnmsmitted, in any form or by any
`means, electronic, mechanical, photocopying, recording, or otherwise,
`without prior written permission from the publisher.
`
`Permission to photocopy or reproduce soleli• for internal or personal
`use is pe rm itted for libraries or other users registe1·ed with lhe Copyright
`Clearance Center, provided that the ba.~e fee of $4. 00 per chapter p l us S.10
`per page is paid clirectli• to the Copi•right Clearance Center, 27 Congress
`Street, Salem, '-'IA 01970. This consent cloes not extend to other kinds of
`copying, such as copying for general distribution, for advertising or
`promotionnl purposes. for creating new collected works, o r for resale.
`
`Printed in the United States of America
`Composition by Graphic World, Inc.
`Project management by Graphic World Publishing Services
`Printing/binding by Maple-Vail Book Manufacturing Croup
`Mosby-Year Book .. lnc.
`11830 Westline lndust..-ial Drive
`St. Louis, Missouri 63146
`
`ISBN: 0-8016-7711-4
`95 96 97 98 99
`
`I 9 8 7 6 5 4 3 2 1
`
`West-Ward Exhibit 1043
`Jennings & Linehan 1996
`Page 002
`
`
`
`CHAPTER 15
`
`Renal, Perirenal, and Ureteral Neoplasms
`
`Scott B. Jennings
`W. Marston Linehan
`
`CLASSIFICATION
`Upper urinary tract neoplasms (Table 15-1) nre of
`rennl parenchymal. urothelinl, connective tissue, and
`metastatic origin. Benign renal tumors include nd·
`enomas, oncocytomas, and other tumors derived frorn
`vascular ond connective tissues. Primary malignant
`renal tumors include renal cell carcinomas, urothelinl
`carcinomas of the renal pelvis and collecting system,
`and sn rcomns originating in the kidney. Secondnry ma(cid:173)
`lignant rnnul tumors may involve tlie kidney by direct
`exte nsion or as a result of hematoge nous sp1·end. Ile·
`mnlologic rnalignnncies such as le ukemia and lym(cid:173)
`phoma <.'Ommonly involve the kidney and arc usunlly
`n manifestation of systemic disease. Primary retro·
`peritoneal tumors are usually malignant and include
`lipomns ns well as the various sarcomas. Primary ure·
`teral tumors mny be benign or malignant and include
`fibrocpithelial polyps, inverted papillomas, and the
`more common urothelial malignnncics. The ureters
`may be involved secondarily by direct extension of
`adjacent mnlignancies or, more common ly, from
`blood-bome metastases.
`
`EVALUATION OF RENAL MASSES
`The cvalunlion of renal masses has changed drn·
`mntically with the advent of improved radiologic tech(cid:173)
`niques. Bemuse of tl1e position of the kidney within
`the retro1>eritoneum, in the past renal masses some·
`times rcmnincd undetected until they grew large
`enough lo produce local symptoms. which gencrall)'
`indicated nd"anced disease. Kow, as tlie routine use
`of nlxlominal ultr:L<Onogrnphy (US) 3nd L'<>mputcrizcd
`tomography (C1) has increased, renal masses are oRcn
`detected at an earlier stage. A systematic method is
`necessary to ensure complete evaluation of suspected
`renal mnsscs. as each radiologic modality ha.< its rel(cid:173)
`ative strengths and weaknesses (Fig. 15-1).
`Intravenous pyclography (JVP), despite its luck of
`sensitivity nnd specificity, remains the initial diag11os·
`
`J
`
`tic method employed for the evaluation of suspected
`renal m:lSSes because of its familiarity to physicians
`and its role in the evaluation of hemnturia. An fVP
`can detect many reonl masses and provide information
`regarding the location and function of the kidneys;
`however, small masses that do not distort the coUect·
`ing system or periphery of the kidney may not be
`detected. Also, patie nts with con trust alle rgies, renal
`insufficiency, or other conditions may not be candi(cid:173)
`dates for IVP.
`lf a renal mass is found on !VP, ukrnsonography is
`ollen recommended to determine if the mass is solid
`or cystic. The majority of masses detected by !VP
`prove to be simple cysts, and US is quite accurate in
`this diagnosis. Strict sonogmphic criteria for simple
`cysts have been defined and include a smooth cyst
`wrul, a round or oval shnpe without internal echoes,
`and good through transmission with strong acoustic
`shadows posteriorly (Fig. 15-2). If these criteria arc
`met, observation is sufficient in asymptomatic pa·
`tlents. A mass th.it deviates from these findings is
`studied further. Percutaneous needle puncture of
`cysts with aspiration of cystic !luid for cytology and
`biochemical studies and fi 1rthcr injection of the cyst
`cavity with contrast med ium for rndlogmphic evalu·
`ation were ollen pe1for111cd in the past. However, the
`nppearnnce of cytology of the cyst Ruicl and biochem(cid:173)
`ical assays were frequently nondlngnostic (Bosniak,
`1993) except in the case of simple benign cysts, and
`tl1us with modem sonographic techniques these stud(cid:173)
`ies are rarely necessary. Indeterminate or complex
`cystic m:lSSes are better evaluated with CT.
`Solid renal m:lSSes demonstrate varied echogenicity
`on US, and thus the specificity of this modality is
`limited. Angiomyolipomas are very brightly echogenic
`and may be suspected from US. but rcnnl cell carci(cid:173)
`nomas may m imic this appearance. Other solid renal
`mnsses mny be isoechoic or hypocchoic nnd may be
`
`643
`
`West-Ward Exhibit 1043
`Jennings & Linehan 1996
`Page 003
`
`
`
`644
`
`AOUf,1' UROLOCl'
`
`well demarcated or irregular. Because of the lack of
`speci.flcity, CT is often recommended for solid masses
`seen on US (Fig. 15-3).
`Occasionally, masses seen on JVP are not detected
`by US. This may be due to the presence of an isoechoic
`renal mass, patient habitus, or a nor1nal anato1nic vari·
`ant such as fetal lobulation or a hypertrophied column
`of Bertin. In the past radionuclide renal scanning
`was recommended, based on the principle that nor-
`
`TABLE 15-1
`Classilicatlon ol Renal, Perirenal, and Ureleral Twnors
`Renal tumors
`Benign
`Adenoma
`Oncocytoma
`Angiomyolipoma
`Fibromo
`Leiomyoma
`Ju.iaglOmerular 1um0<s
`Hemangk>ma
`Lipoma
`Primary maUgnant
`Renal cell carcinoma
`Urotl>elial carcinoma
`Sarcoma
`Secondary malignant
`Relroperitoneal sarcoma,
`pancreatic carcinoma.
`co!()n carcinoma (d~ecl
`extension)
`lung, breast, GI
`(hematogenous sp<ead)
`Lymphoma, leukeraa
`
`Primary retroperitoneal tumors
`Benign
`Upoma
`Lelomyoma
`Cysts
`Malignant
`~osarcoma
`Fibrosarcoma
`Lelomyosarooma
`Ureteral tumors
`B<mign
`Papllloma
`Fibroepithelia1 polyps
`Primary malign0S1t
`Urothellal carcinoma
`Sat coma
`Secondary malignam
`Retroperltoneal sarcoma,
`lymphoma, breasl, GI,
`ceNlx, prostate
`
`FIG. 15-1
`Algorilhm for evaluation ol a renal mass.
`
`mal function ing renal tissue is necessar)' to conc-en(cid:173)
`trate commonly used n1dioisotopes (Newhouse, 1993},
`and thus tumors or cysts appear as areas of de(cid:173)
`creased activity, and columns of Bertin are evident as
`areas of increased intensity (Fig. 15-4). However,
`contrast CT seu1s C<lll often readiJ)' make these diag(cid:173)
`noses and [n addition provide much more infonnation
`regarding tumor staging. Thus CT is c<msidered by
`most to be the moda Uty of choice in this situation,
`although n uclear renal scans are still comm only per(cid:173)
`formed.
`CT scanning is the predominant method used today
`in the evaluation of renal masses and has largely re(cid:173)
`placed renal angiogmphy in this regard. The advan(cid:173)
`tages of CT scanni.ng include greater sensitivity in
`detecting small renal masses, imp1-ovcd detection of
`small amounts offilt within tumors, the abi lity to quan(cid:173)
`titate the density of VRrious components, and more
`accurate tumor staging because of improved visual(cid:173)
`ization of the retroperitoneal sb·uctu.res, adj:icent or(cid:173)
`gans, and associated vasculature. In addition , CT scan(cid:173)
`ning is considered less invasive and less expensive
`than arteriography. CT scanning is most often per(cid:173)
`formed with and without oral and intravenous contrast
`administration to fully realize the diagnostic potential
`of this modality (Bosniak, 1993); thin sections (5 mm
`or less} may also be important (Fig. 15-5). In a small
`percentage of patients, CT will not be able to differ(cid:173)
`entiate atypical angiomyolipomas, complex cystic
`masses, and other inAammatory or infectious masses
`from more typical solid tumors, and further evaluation
`\Vith mngoetic resonan<.'e imaging, renal angiography,
`
`IVP
`
`I
`~ Ullm101111d~
`
`Solid or
`
`No Mass
`
`Simple Cyst
`I
`Observe
`
`Com,pl/ x - - -
`Nuclear Renal Sct1n
`I
`Pse-ud~tumor
`~CT...............
`~ Observe
`~ /
`Fa1
`Solid
`lnde1erminn1c
`I
`Vascular
`/MRI
`
`Tumor
`I
`Surgery
`
`~ l11C'01l,lusivc ~
`
`/
`/ Aneriogrom
`Vnsculor
`Benign
`I
`I
`Tumor
`Observe
`I
`Surgery
`
`FNA ............
`
`Surgcl)'
`
`~1alignan1
`I
`Surgery
`
`--
`
`West-Ward Exhibit 1043
`Jennings & Linehan 1996
`Page 004
`
`
`
`' -. '
`
`SGS
`
`RE.>1111., PF.RlllEi'IAL, MW URETER.AL NF.OPW\S.\/S
`
`645
`
`'
`
`FIG. 15·2
`Ullrasouncl ol a simple renal cyst showing renal parenchyma
`(short arrows), cysl wall {long arrows), and slrong poslerlor
`wall (arrowheads).
`
`FIG.15-3
`Ultrasound examination ol a solid 1enal mass (arrows).
`
`FIG. 15-4
`Olmercaptosuccinic acid (OMSA) scan of a renal pseudolu·
`mot (hypertrophied column ol Ber11n) shown by arrows.
`
`-
`
`West-Ward Exhibit 1043
`Jennings & Linehan 1996
`Page 005
`
`
`
`646
`
`Aou1.r Unoux;r
`
`AG. 1S-5
`CT of a renal cen carcinoma (RCC) (arrows). A, Without contraS1 enhancement. B, With contrast enhancement (From Wiiliams
`RO, Tanagho EA: Current Surgicel Diagnosis and Treatment. Los Altos, Calif. Lange Medical Pub4ications. 1985, p 860. Used
`by permission.)
`
`fine needle aspiration cytolo1,'Y. or ultimately surgical
`exploration may be necessary.
`In nddition to its role in the evnluation of indeter(cid:173)
`minnte renal masses, renal :trterlography is also useful
`in dellning the blood supply to large tumors, or for
`precise delineation of the rennl vasculnture when
`nephron-sparing surgery is contemplated. Arteriog(cid:173)
`raphy is not without risk, and digital subtraction an(cid:173)
`giography in combination with CT has been proposed
`as an nltemative to decrease morbidity as we ll as give
`acceptable anatomic definition ('Zabbo et al., 1985}.
`Magnetic resonance imaging is an increasingly im(cid:173)
`portant modality in the diagnosis and staging of renal
`masses and offers sign ificant advantages over CT in
`some respects (Fig. 15-6). The teclmiquc Is particu(cid:173)
`larly useful in the evaluation of vascular invasion and
`involvement of adjacent organs. The addition of gado(cid:173)
`linium contrnst administration to MRJ has greatly in(cid:173)
`creased its ability to detect renal masses smllller than
`3 cm, as well as assisted in the diagnosis of true in(cid:173)
`vasion of lbe wall of the venal ca.va (McClennn.o and
`Deyoe. 1994). TI1e evaluation of local adenopathy is
`also impro"ed with MRI (Semcll-3 et al., 1993). ln
`addition, MRJ with gadolinium Is possible in patients
`in whom scanning with ioclinntecl contrast agents is
`contraindicated. At tlie present time MRI is perhaps
`most useful as an adjunct to contrnsl CT scans, partly
`because of greater cost, bul il undoubtedly will play
`a lnrger role in the future.
`
`BENIGN RENAL TUMORS
`Benign tumors of the kidney are common and in
`the past were detected primarily at autopS)'. as the
`majority arc asymptomatic. However, a growing num-
`
`AG. 15-6
`TransaxialT,-welghted MRI of Iha same RCC (arrows) shown
`In Fig. 15·5, A and B. (From Wiiiiams RO, Tanagho EA:
`Current Surgical Diagnosis and Treatment. Los Altos. Calif,
`Lange Medical Publications, 1985, p 860. Used by perrnls·
`slon.)
`
`bcrofthese generally small masses are being detected
`ns the routine use of abdominal sonogrnphy nnd com(cid:173)
`puterized tomography has inc1·eased. Often there are
`no clear-cul clinical or rndiologic features lo CQnfirm
`the diagnosis preoperntivcly, and thus the pathologist
`determines the tnae etiology of these tumors followir>g
`surgery.
`
`Ade no ma
`Renal C<>rtical adenomas are benign tumors that arc
`usually found at autopsy because of their small si7.e
`
`West-Ward Exhibit 1043
`Jennings & Linehan 1996
`Page 006
`
`
`
`R£NJ\t, PEltl/lfNM,, AND URffelv\L Nf:ON .. ~SMS
`
`647
`
`FIG. 15-7
`Histologic section of true renal adenoma (original magnifi(cid:173)
`cation, x 100).
`
`FIG. 15-8
`Histologic section of a benign renal oncocyloma (original
`magnification, x 200).
`
`and gene rally asymptomatic clinical course. They have
`been noted in 7% to 22% of autopsy specimens (Bon(cid:173)
`sib, 1985). Adenomas are usually less than 1 c m in
`size and located exclusively in the cortex, with a dis(cid:173)
`tinctive g1-ay-white or yellow appearance on gross ex·
`amination, without necrosis or hemorrhage (Fromo(cid:173)
`wit-l and Bard, 1990). Histologically they are com(cid:173)
`posed of eosinopb ilic or basophilic cells in a papillary
`or tubu lopapiUary patte rn (Fig. 15-7), typic.~lly sepa(cid:173)
`rate fi-om the surrounding parenchyma and usually
`without a capsule (O'Toole et al., 1993). Any evidence
`of clear cells, mitoses, nuclear pleomorphism, or ne(cid:173)
`crosis obviates the diagnosis of adenoma.
`The e tiology of adenomas is unclear, although they
`have been associated with s moking, arte riolar neph(cid:173)
`rosclerosis, and dialysis (Mosto6 et al., 1988). Histo(cid:173)
`chemical and ultrastructuntl s tudies point to a distal
`tubular origin for adenomas (Fromowitz and Bard,
`1990). Other investigators (Kovacs, 1993) have noted
`the chromosomal changes of trisomy 7, trisomy 17,
`and loss of the Y chromosome. M e loni and colleagues
`(1992) found similar changes in adenomns b ut caution
`that similar or identical changes are found in tumors
`that histologically prove to be renal cell carcinomas.
`Controversy pe rsists in the literatu re regarding the
`true clin ical nahire of these neoplasms. In an earl)'
`autopsy series reported b y Dell (19.50), tumors less
`than 3 cm in size demonstrated a very low rate of
`metastasis, and by conven tio11 renal masses smalle r
`than 3 cm were referred to as adenomas. Howeve1",
`multiple reports, including Bell's original series, cloc(cid:173)
`tunen t me lastases originating from small tumors, and
`thus size a lone is not an accurate predictor of malig(cid:173)
`nancy (Donsib, 1985; C urry ct al., 1986; Fromowitz
`nad Bard, 1990). Bennington and Beckwith (1975) con(cid:173)
`cluded that adenomas could not re liably be differen(cid:173)
`tiated from small aclenocarcinomas. However, oth er
`
`investigators feel that histologio criteria can be used
`to distinguish these two neoplas ms (Bonsib, 1985; Fro(cid:173)
`mowitz and Bard, 1990; Mostofi e t al., 1984; O'Toole
`e t :ti. . 1993). Because it is impossible to differentiate
`adenomas Ii-om carcinomas preoperatively on the basis
`of symptoms, size, or radiographic appeai-ance, small
`solid renal ma.~ses should be conside1·ed to be malig(cid:173)
`nant until pl'oven otherwise.
`
`Oncocytoma
`Renal oncocytomas are benign renal tumors com(cid:173)
`posed of oncocyte.~. a cell population with intensely
`eosinophilic granular cytoplasm . The y were first char(cid:173)
`acte rized as a clinical e ntity by Klein and Vale nsi
`(1976) and accoun t for 3% to 7% of solid renal tumors
`(Lieber, 1993). Grossly, oncocytomas have a typical
`tan or mahogany t'Olor and are well circumscribed with
`a fibrous capsule. invas ion into adjacent pare nchyma,
`collecting syste m, or renal capsule is rare. On cross(cid:173)
`section, the tumor appears to he homoge neous with(cid:173)
`out hemorrhage or necrosis. A central stellate scar is
`ofte n present, particularly in la1-ger tumors. Oncocy(cid:173)
`tomas nre generally un ilateral, although approximately
`6% of patients have bi lateral tumors (Lieb er, 1986)
`and isolated cases of multifocal tumors have been re(cid:173)
`ported (Warfel and Eble, 1982). Oncocytomas may
`become quite h rge, with a median diameter of 6 cm
`in collecled series. Oncocytomas are also found in
`multiple other organs, including the thyroid and para(cid:173)
`thyroid glands, the adrenal glands, and the salivary
`glands.
`H islologic,1lly, oncocytomas are composed of large
`polygonal cells with nn intensely eosinophilic granular
`cytoplasm (fig. 15-8). Ultnistructural studies have
`shown that the granularity is clue to abundant mito·
`chi;mclria within the cytoplasm. Mitoses and other cel(cid:173)
`lular organelles ni·e rnl'e, as is nuclear pleomorphism.
`
`West-Ward Exhibit 1043
`Jennings & Linehan 1996
`Page 007
`
`
`
`648
`
`ADULT UROl.OCr
`
`The exact cell of origin is unknown, although recent
`studies suggest an origi11 from the iotercalated cells of
`the collecting ducts (Storkel et al., 1989; Zerbnn et
`al., 1987). Liebe1· ancl associates (1981) pr'Oposed a
`grading system for onc'OC)'toma.~. with grade 1 tumors
`composed of well-differentiated cells with regu lar nu(cid:173)
`clei and abu ndant cytoplasm , grade 2 tumors dem·
`onstrating more variegation of the nuclei a11d cyto(cid:173)
`plasm, and grade 3 tumors degrading juto significant
`nuclear pleomorphism with possible mitotic activity.
`The term oncocytomo s hould be used to refer only to
`those tumors composed of a pure population of on(cid:173)
`cocytes, i.e., very well differentiated eosinophilic
`granular cells. Many renal cell carcinomas also con tain
`granular cells alone or in combination with clear cel ls
`or spindle cells, and on occasion pathologic differen(cid:173)
`tiation of oncocytoma from carcinoma may be difficu lt.
`However, the granular cells found in these carcinomas
`are less well differentiated and demonstrate obvious
`nuclear p leomorphism.
`Clinically, most oncocytomas are asymptomatic, de(cid:173)
`spite their occasional large size. Most tumor< are dis(cid:173)
`covered incidentally, although occasionally patierlts
`present with flank or abdominal pain, microscopic or
`gross hematuria, or a palpable mass. They are ap(cid:173)
`proximately twice as common in males as in females,
`similar to series of renal cell carcinomas. Oncocytomas
`most commonly occur in the early seventh decade,
`somewhat later than re1ial cell carcinomas (Lieber,
`l993).
`Radiologically, oncocytomas appear to be typic'.tl
`solid renal mass lesions. There is no pathognomonic
`finding on intravenous pyelography, renal ultrasonog(cid:173)
`raphy, or comp uted tomography, although a central
`stellatc. scar can be visualized with sonography or to·
`mography in some cases. However, th is finding is non(cid:173)
`specific. \Veiner and Bemstein (1977) noted several
`characteristic findings on renal angiography of onco(cid:173)
`cytomas and used the term "spoke-wheel" to describe
`the appearance of the feeding artelies. These 6 ndings
`are not always present, however, and may also be seen
`in re nal cell carcinomas (Morra and Das, 1993).
`Flow cytometry studies of renal oncocytomas have
`)~elded conllicting data. Rainwater and colleagues
`{1986) found aneuploid or tctraploicl histograms in 50%
`of their specimens. Other investigators have re ported
`aneuploidy or tetraploidy in 0% to 4% of patients
`(Hartwick e t al. , 1992; Psihramis and Golclber·g, 1991;
`Veloso et nl. , 1992). The p revalence of aneuploicly or
`tetraploidy did not correspond with an unfavorable
`clinical course. and thus flow cytome try does not seem
`to be useful as a prognostic indicator. Cytogenetic
`studies also fail to show consistent characteristic ab(cid:173)
`nonnalities (Kovacs, 1993).
`Since true oncocytomas are almost without cxcep·
`tion clinically benign, nephron-sparing surgery has
`
`been proposed in selected cases. This treatment prc(cid:173)
`mmes fill accurate preoperati"e diagnosis, particu larly
`the excl usion of renal cell carcinoma. As pre\fiC>usly
`noted, there are no completely reliable clinical 01· ra(cid:173)
`diographic criteria to distinguish oncocytomas from
`carcinomas. Needle aspiration cytology has been sug(cid:173)
`gested as a diagnostic modality for patients in whom
`oncocytoma is suspected. Rodriguez and associates
`(1980) reported the Srst case of oncocytoma diagnosed
`preoperatively by aspiration cytology, with Anal di(cid:173)
`agnqsis confirmed following total neplueetomy. Othe1·
`small series (Cochand-Priollet et al. , 1988; Carcia(cid:173)
`Bonafe et al., 1993; Gupta et al., 1990) have collRrmed
`the potential usefulness of preoperative aspiration cy(cid:173)
`tology, but in these series cytologic criteria could not
`exclude carcinoma in 20% to 33% of patients. If nepb(cid:173)
`ror1-sparing surgery is contemplated, Rne-needle as(cid:173)
`piration cytology may be helpful, but thorough tumor
`sampling must be performed and results should be
`interpreted with caution, since carcinomas may con(cid:173)
`tain foci of oncocytic cells and appar·ent pure onco(cid:173)
`cytomns may harbor small areas of malignant degen(cid:173)
`eration. Also, the possibility of multifocal tumors must
`be considered. If the diagnosis of oncocytoma can he
`unequivocall)• established prior to surgery, a nephron(cid:173)
`sparingapproach may be indicated. However, because
`of the inherent uncertainties in current preoperative
`diagnostic modalities, most patients with oncocytoma
`utimately undergo racLcal nephrectomy.
`
`Anglomyolipoma
`Renal angiomyolipomas are benign tumo1·s ofien
`associated with tuberous sclerosis, nn autosomal dom(cid:173)
`inantly inherited disorder characterized by mental re(cid:173)
`tardation, epilepsy, adenoma seba~-emn, and hamar·
`tomas of the brain, retina, heart, bone, lung, and
`kidney (Stillwell et al., 1987). As many as 80% of pn(cid:173)
`tients with tuberous sclerosis will ultimately develop
`renal angiomyolipomas, and in these patients the tu·
`mors are·generally small, bilateral, and asymptomatic.
`Renal cysts also common ly occur in patients with tu(cid:173)
`berous sclerosis. The majority of renal angiomyoli(cid:173)
`pomas are not associated with tuberous sclerosis. 1n
`recent series only 17% to 20% of patients with an·
`giomyolipomas also demonstrated evidence of tuber(cid:173)
`ous sclerosis, and thus it does not seem like!)' that
`isolated renal angiomyolipoma represents a forme
`fruste of tuberous sclerosis (Bret et al., 1985; Steiner
`et al., 1993). ln these patients the tumors are unilal·
`era!, usually oecur in women 40 to 60 years of age,
`and are often large and symptomatic.
`Angiornyolipomas nre composecl of generally m:i(cid:173)
`ture fat cells, smooth muscle, and abnorm<tl blood
`vessels. Sometimes one cell type may predominate.
`C rossly. the tumors are yellow to gray, depe11ding on
`the proportion of fat nnd muscle ~-ells. 11iey a1·e unen-
`
`West-Ward Exhibit 1043
`Jennings & Linehan 1996
`Page 008
`
`
`
`RENAL. P£R/l!E:/\'M., AND UllETEMI. N~:OPU.SlJS
`
`649
`
`capsulated 3lld may extend into t11e collecting syste m
`or perirenal fat. Hemorrhage nnd necrosi.'l nre com(cid:173)
`mon . Histologicnlly, the adipocytcs are usually mature
`nnd regular in size (Fig. LS-9). The blood vessels nre
`thick-walled and tortuous. The smooth muscle cells
`mny be spal'se or may form sheets of cells that muy
`be confused with leiomyoma (MostoR and Davis,
`1984). Retroperi toneal lymph nodes and adjacent or·
`gnns such as the liver or spleen may sometimes contain
`nngiomyolipomas identical to the renal tumor, and this
`Is felt to represent multifoeallty rather than aggressive
`metastatic behavior, as no distant metastases have
`been reported from pure nngiomyolipomas. However,
`malignant degeneration of nngiomyolipoma into snr·
`coma with d istan t metastiL•cs has been reported (Ferry
`et nl. , 1991; Lowe et nl., 1992), and in these instances
`the outcome was fatal. Angiomyolipomas and rcnnl cell
`
`FIG. 15-9
`Hlstologic: section or a renal anglomyolipoma showing typlc:at
`lat (short arrows), muscle (loog arrows), and vessels (arrow(cid:173)
`heads) (original magnification, x 63.)
`
`carcinomas may also coexi.'lt in the same kidney (Kn·
`vnney and Fielding, 1975; Schujman et al., 1981; Tak·
`cyamn et al., 1982; Silpannntn et al., 1984).
`The clinical prescntntion of angiomyolipomM is
`vuriable and appears to he related to the size of the
`tumor. Steiner and associates (1993) found thnl no
`pt1tle nt with a tumor less thnn 4 cm in si1.e was sym1>·
`tomntic. Tumors larger Lhan 4 cm in size are symp(cid:173)
`tomatic in 46% to 82% of cases (Oesterling et al. , 1986:
`Steiner et al., 1993). Flank or abdominal pain is the
`most common presenting symptom, occurring in ap(cid:173)
`proximately 50% of patients, followed by p.-tlpable
`mass and hemnturia. Hypertension and anemia nre
`also common. Patients mny also present in hemor·
`rhngic shock following rupture of the tumor into the
`kidney or retrope ritoneum.
`Bnd lographic studies arc usually he lpful in the di·
`agnosis of angiomyolipomn. Intravenous pyclography
`cannot dilTere ntiate angiomyolipoma from other solid
`renal lesions, and the nngiographic findings typically
`seen are also common in renal cell carcinoma. How·
`ever. ultrasonogrnphy and CT scanning can reliably
`define angiomyolipomn in the majority of cases. An·
`glomyolipomas are the most echogenic renal masses
`found on sonogrnphy duu to the numerous fat-nonfat
`interfoces seen in most tumors (Pitts et al. , 1980).
`O the r renal musses nrc, however, occasionally highly
`cchogenic, includ ing a smoll proportion of rennl cell
`carcinomas. The presence offal densities(- 70 lo - 30
`Hounsfield units) on CT scanning is almost pathogno·
`monic for angiomyolipoma (Fig. 15-10). Detection of
`small amounts of fat within the tumor may therefore
`be critical for diagnosis , nnd such detection may be
`improved by using noncnhnnced scans with LS to 5·
`mm sections (Bosnink ct nl ., 1988; Kurosaki et ul.,
`l993). Magn etic resonance lmnging may also be useful
`
`B
`
`FIG. 15·10
`CT of a renal angiomyolipoma. A , Typical mass le51on with internal areas of fat (arrows). B, Same lesions as In A showing
`abundant lat (arrows).
`
`West-Ward Exhibit 1043
`Jennings & Linehan 1996
`Page 009
`
`
`
`650
`
`ADULT UROLOCl'
`
`because or the characteristic higb signal intensity of
`fat on T,-weighted imogcs. All these diagnostic mo·
`dalities are dependent upon the relative proportion of
`fat within the tumor. and thus a small number of cases
`will remain of uncertain etiology following complete
`radiologic evaluation. Fine-needle aspiration biopsy
`has been recommended in such instances with good
`results (Snnt et al. , 1990: Toavitsainen ct nl., 1989).
`The treatment of angiomyolipoma is dependent
`upon the size of the tumor, the presence of symptoms.
`:wd the accuracy of diagnosis (Andriole, 199'2). Small,
`asymptomatic tumors discovered incidentally may be
`followed with )'Carli• sonography if the diagnosis is
`unequivocnl. Asymptomatic patients with tumors
`larger than 4 c m may still ho candidates for observation
`with :mnunl or semiannual ultrasound, but nenrly 50%
`of these tumors will grow (Steiner et al .. 1993). Tumor
`embolizMion or nepbron-sparing surgery should then
`be considered to pre"cnt further loss of renal paren·
`chyma. Symptomatic pntionts usunlly requin: treat·
`menl, regard less of the size of the muss. Embolizalion
`or parlinJ nephrectomy arc the preferred modes of
`therapy, although totnl nephrectomy will be indicated
`in some cases. Patients who present with life-threat(cid:173)
`ening hemorrhage should undergo immedinte surgical
`exploration. Radic.,l ocphrectomy is the trcntment of
`choic'<l, although partinl nephrectomy 1nai• be t'Onsid·
`ered if the clinic.,J situation allows and if angiomyoli·
`poma is suspected preopcmtively and confirmed by
`intraopcmtive frozen-section biopsy. Finally, patients
`io whom the mdiologic diagnosis is uncerlllin should
`undergo flne·needle aspiration biopsy or surgical ex·
`ploralion, ns for any other solid re mtl mnss.
`
`Other Benign Tumors
`Tumors ha"e been reported to arise from virtual!>'
`all of tl1e diverse cell types of which the kidney is
`composed. Medul.lary interstitial fibromas nre small
`benign tumo1·s commonly found at autopsy, occurring
`in 18% to 40% of patients (Reis e t al.. 1988; Warfel
`and Eble. 1985). As the name implies, these tumors
`arise from the interstitial cells nf the renal medulla
`which are belie"ed to cau.se an antil1ypertensive effect
`(Lerman et al., 1972). However, no convincing e\'i·
`dence exists that interstilinl fibromas arise as n resu lt
`of hypertension (Stuart ct nl., 1976). The tumors are
`genemlly small, white or gray in color, and round or
`oval in shn1>e. The majority are multifocal. They occur
`in males and females \\1th equal frequency (Eble,
`1990) and arise as early as the second decnde of life
`without increasing signiftcu11tly in frequenc)' with ad·
`vanccd age (Reis et al., 1988). H.istologicnlly, the tu·
`mors arc composed of stellate or spindle-shaped cells
`loosely arranged in a basophilic or hyalini1.ed matrix,
`with atrophic renal tubules nt the periphery. :.tost of
`these tumors are clinically silent, although occasion·
`
`ally they may cause flank or abdominal pain or he·
`maturin. The few symptomatic cases have genernll)'
`been h'eated with nephrectomy, although ncphron·
`sparing surgery should be considered if the diagnosis
`Is suspected preoperatively.
`Leiomyomas are rare benign tumors found in n1)·
`proximately 5% of patients undergoing autopsy (Xi·
`pell , 1971). They nrc almostalwnys small, subcapsular,
`and n.•ymptomatic. The occurrence of multiple tumors
`Is common. These tumors arise from the renal capsule,
`"essels, or, rarely, the renal pelvis. Leiomromas are
`gray or white in color, usually sm:iller than 1 to 2 cm,
`nnd usually well encapsulated (DI Palma and Giardini,
`1988). Histologicnll y they are composed of fusiform
`smooth muscle cells ttrranged in an interlaced fashion.
`Mitotic 6gures are rare or absent. A much smaller
`subset of patients present with large, solitary, symp·
`tomatic tumors. A palpable abdominal or Hank mass
`is the most common sign at presentntion, followed by
`nbclominal or Annk pain and he mnturia (S teiner cl nl.,
`1990). Clinically apparent leiomyomas arc more co111-
`mon in young white women. Total nephrectomy is
`usually performed because of the uncertainty of di·
`ngnosis, but renal-sp.1ring surgery is advocated if tech·
`nicnlly feasible and if the diagnosis can be conRm1ed
`lntrnoperatively.
`Ronin-secreting juxtaglomcntlar cell tumors arc
`mre but important benign renal masses, with fewer
`thnn ·10 cases reported in the literature. Patients with
`these tumors present with severe diastolic h;"j>Crten·
`sion, hypoJ..-nlemin, and ele"atcd plasma renin levels.
`Typically, patients :ire yow1g (70% under 25 years of
`age) 11nd 66% nre female (McVlcnr ct al. , 1993). Head(cid:173)
`ache is the most common presenting symptom. The
`diagnosis is usually suspected from the finding of hy·
`pcrtension in a young patient and is confirmed by
`elevated renin levels and secondary hyperaldoster·
`onism and hypolmlemia, :ilthough these secondary
`findings are not always present. Compute rized to·
`111ography is the most reliable imaging method to dem·
`onstmte these tumors. Renal angiography is some·
`times useful, primarily
`to exclude renovascular
`hypertension. Selective renal vein rcnin sampling may
`nlso be he lpful , although this is diagnostic in fewer
`th1111 50% of cases (McVica1· el al., 1993). Crossly, the
`tumors are gener;1lly smaller thnn 3 cm, grayish-ye!·
`low in color. and well encapsulated. Histologically,
`they originate from the smooth muscle medial ce