BULLETIN OF

THE NEW YORK ACADEMY

OF MEDICINE

VOL. 30, NO. 12 @ DECEMBER 1954

THE INFLUENCE OF CLINICAL ANDLABORATORY INVESTIGATION INESTABLISHING CURRENT THERAPY

OF. BLADDER CANCER*

ROGER BAKERProfessor and Director of the Department of Urology,

Georgetown University School of Medicine, Washington, D.C.

3jH E Genito-Urinary Section of The New York Academy5 of Medicine had its first meeting on December 9, 1890.

E T g The first paper was read by Doctor F. R. Sturgis, en-titled "The Diagnostic Value of Hematuria in Affections

essz~z5R5255 of the Genito-Urinary Organs," which undoubtedlyhad a direct bearing on our topic this evening. Since that time some ofthe most significant contributions to the problem of bladder cancerhave been made by urologists from the New York scene. To mentiononly a few, we are all familiar with the work of Buerger, Keyes, Beer,Barringer, McCarthy, Dean, and more recently, the outstanding workof Marshall and Whitmore.

ETIOLOGY

Concerning the cause of bladder cancer little is known, but it is* Presented by invitation before the Section on Genito-Urinary Surgery of The New York Academy

of Medicine, Dec. 16, 1953. Manuscripts received April 1954.This work supported in part by an American Cancer Society Institutional Grant.

920 THE

relatively great when compared to the paucity of etiological factorsresponsible for other major cancers of man. Rehn in 1895 first reportedbladder cancer in fuchsin workers. Hueper proved that aniline dyeworkers have a significantly greater incidence of cancer than in theaverage population. Experimentally, bladder cancers have been producedby 20-methyl cholanthrene and B-naphthylamine. Acetylaminofluorineand several other azo dyes also selectively produce carcinoma of thebladder. It is interesting at this point to return to history. In i870,Billroth, the famous Viennese surgeon, contended that "without pre-vious chronic inflammation, cancer does not exist." In I907, Ribbertstated that cancer originates at the site of local alterations in tissuemetabolism. In point of fact, we know that cancer of the bladder hasoccurred in patients with Hunner ulcers, diverticula, and bladdermetaplasia from whatever cause. Leukoplakia of the bladder is con-sidered a premalignant lesion. Benign papillomas of the bladder invari-ably become malignant if untreated. There is a 2 per cent incidenceof malignant degeneration occurring in exstrophy of the bladder. Manypatients with bladder cancer have a vesical calculus. In these lattercases the tumor itself is invariably squamous cell and frequently sur-rounded by marked areas of hyperplasia or metaplasia. Was the hyper-plasia or metaplasia first, and the malignant degeneration second, orvice versa? The problem remains unanswered, though speculation isrampant and to say the least, intriguing. In our own laboratory we areconcerned to some degree with this phase of bladder cancer. B-naph-thylamine fed to dogs produces bladder cancer in a minimum of elevento twenty-four months. If a foreign body, such as a piece of wood ora calculus, is placed within the dog's bladder or a surgical exstrophyproduced, and then B-naphthylamine administered, it appears that theaverage time for development of cancer is only six to twelve months.In all cases hyperplasia or metaplasia apparently precedes developmentof the tumors. While we have no evidence that "chronic inflammation"'is carcinogenic per se, preliminary evidence indicates that it acceleratesthe development of bladder cancer in the organism disposed towardtumorigenesis.

CLASSIFICATION OF BLADDER TUMORS

While many different classifications of bladder tumors have beendescribed, covering the range from fibroma to hydatid cyst to fibro-

9 2 0 THE BULLETIN

921

myxosarcoma, from a practical standpoint three tumors comprise over97 per cent of those encountered in practice: transitional cell, squamouscell, and adenocarcinoma. The present paper is concerned only withthese cellular types. It should also be mentioned that papillomas arenot included in this report as these are benign, though premalignant,lesions.

Biopsy: Prior to the investigative work of Jewett and Strong' itwas common practice to biopsy bladder tumors by means of endoscopicsnares or forceps. The contribution of these men and the confirmingdata of Marshall and others have demonstrated that depth of infiltrationis more important in determining therapy and prognosis than gradingof the cancer cells. Potentially, cancers that have infiltrated more thanmidway the bladder muscle will have spread widely and prognosis isextremely poor (about I5 per cent cures) regardless of the acceptedmethod of therapy employed. Potentially, about 85 per cent of patientswill be cured, regardless of accepted type of therapy utilized, who havecancers that have penetrated to only the superficial muscle layers orless. Grading of tumor cells affords some indication of rate of growth,but frequently is deceptive. A tumor may be classified as Broder gradeone at the surface but be grade three at the base. A grade one or twocancer may have spread through the entire bladder wall and be widelydisseminated through lymphatics. On the other hand, a grade four,highly anaplastic cancer may involve only mucosa, submucosa andsuperficial muscle layers. In the modern treatment of vesical neo-plasms, therefore, the endoscopic snares and forceps must be discarded.All patients (even those with benign papilloma) must have transureth-ral resection of a segment or all of the tumor for biopsy purposes.

Birnazatrl Examination: Some urologists have stated that bimanualexamination is of assistance in determining operability of bladder cancer.Based on observations with some 69 cystectomies, I have been unableto obtain much additional information by this method in the patientsin whom such data were most needed. A routine deep transurethralresection biopsy and frequently laparotomy, are the usual procedures,in my experience and that of many others, which are necessary todetermine operability and provide a reasonable evaluation of prognosis.Perhaps increasing the number of cases in this series will demonstratea greater dependence on this highly subjective examination.

Exfoliative Cellular Pathology: In i892, Ferguson was one of the

Bladder Cancer 9 2 I

922 THE BULLETIN

first to draw attention to the importance of repeated microscopicalexaminations of the urine in the diagnosis of bladder tumors. He ex-amined small fragments of tissue in the debris of the urine collected.Young restressed the importance of this technique. The present-dayuse of the Papanicolaou smear technique is significant when positive,but a negative result certainly fails to exclude the diagnosis of bladdercancer. From the financial and technical 'aspects it is impractical atthis time to use this procedure as a screening test. In our patients wehave discontinued entirely use of this technique for the followingreason: If the Papanicolaou smear is positive, cystoscopy is certainlyindicated; if there were sufficient signs or symptoms of bladder tumorto warrant collecting urine and performing a Papanicolaou smear andit proved negative, cystoscopy is still indicated.

TREATMENT OF INFILTRATIVE VESICAL CANCER

Once the depth of infiltration of the bladder cancer has beenestablished by transurethral biopsy, what type therapy should beemployed: radiation, segmental resection, cystectomy, transurethralresection with or without insertion of radon seeds, or some combinationof these procedures? Unlike Don Quixote who went around tiltingat windmills, urologists are practical specialists requiring a definitiveplan of attack to this problem. The greatest percentage of cases seenby the urologists already have far-advanced, so-called "inoperable,"cancer that has spread beyond the bladder and extensively involvesthe lymphatics, or metastases are known to be present. Statisticallyspeaking these patients are managed best by palliative measures tocontrol bleeding, urinary signs and symptoms, or uremia. This treat-ment consists of transurethral resection of the tumor with or withouturetero-intestinal or ureterocutaneous anastomosis. Most urologists arein agreement with this program. Due to production of distressingvesical symptoms, few American urologists use only radiation therapyfor patients comprising this group. One school of urologic thoughtcontends that palliation in these patients is a defeatist attitude, andis in the process of exploring extremely radical pelvic surgery todetermine whether or not it has any merit. As the results are not yetfully complete it is unwise to present definitive comment at this time.These surgeons are accumulating valuable data which will inevitablyincrease our knowledge of management of this disease.

THE BULLETIN9 2 2

-~~~~~ ~lde Cacr

TREATMENT OF NON-INFILTRATIVE VESICAL CANCEROnly a small percentage of patients with vesical malignancy are

seen by the urologist while the cancer is still confined to the miucosa,submucosa or superficial muscle layers of the bladder. None the less,this small group of patients have provoked the greatest controversyamong urologists specializing in this field. It would appear advisableto discuss the rationale for surgery in cancer: As yet we have no curefor cancer other than radical surgery and, in some cases, radiationtherapy. Curative surgery for cancer is based on the principle that ata particular time in its growth the cancer has not spread beyond thereach of the scalpel. The goal is to operate before the spread, whilethe tumor is still localized. Concerning bladder cancer this logicallyimplies cystectomy. Most urologists contend, however, that the adverseeffects of diversion of the urinary stream after cystectomy are fraughtwith so many complications that transurethral resection or segmentalresection of the tumor are procedures of choice. In addition, it iscontended, cystectomy fails to cure any more patients with cancerthan these "conservative" procedures. This misconception has beenbased on statistics in which cystectomy was used generally as a desper-ate, heroic or "last ditch" procedure for far-advanced bladder cancer.In the majority of these reports cystectomy was resorted to only aftermore conservative treatment had failed; as such, they must also beclassified as failures for conservative management and this is seldomspecified. In addition, only a few of these reports stated whether ornot any lymphadenectomy was performed in combination with cystec-tomy. It is understandable that these statistics offer such a poor rateof cure. Only a few reports have been published of cystectomyutilized exclusively for early or non-invasive bladder cancer. As maybe expected, the results in the latter group are considerably moreencouraging. While the complications of diversion of the urinarystream are real, pro tem, consider the subject from the standpointof good and bad cancer surgery. A thoracic surgeon would be subjectto criticism if he encountered a carcinoma of the lung and removedonly the cancer with a small cuff of normal tissue surrounding it andfailed to remove the rest of the lobe of the lung. A general surgeonwould be quite remiss, if operating on carcinoma of the lower recto-sigmoid he simply excised the malignancy with a cuff of normaltissue and closed the hole in the bowel wall. There are countless

Bladder Cancer 9 2 3

924 THE

similar examples pertaining to any surgical discipline treating cancer.The same urologist that rightly insists upon nephro-ureterectomy fora small cancer of the renal pelvis will perform local excision of abladder cancer surrounded by a centimeter or two of normal bladderwall and then close the defect in the bladder wall. Another urologistwill treat the same non-infiltrative bladder cancer by transurethralresection. The surgeon resects only that portion of the bladder thatendoscopically appears to be involved with the tumor. Obviously,this norm is too subjective and entirely inadequate in many instancesas we have demonstrated by deliberately performing total cystectomya week or two following what appeared to be complete transurethralresection of the tumor. Second, the operator is restricted in his resectionby fear of perforation of the bladder. Third, no lymphadenectomy isperformed to remove the positive tumor bearing lymph nodes thatoccur in some 2 to 5 per cent of these cases. It would appear thatgenerous segmental resection and lymphadenectomy or total cystectomyand lymphadenectomy obviate these objections and offer to date thebest opportunities for cure in treatment of non-infiltrating bladdercancer. The fact that many non-infiltrating vesical neoplasms arecured after transurethral resection as now performed should not satisfyor placate the urologist. We are interested in improving our presentstatistics.

Hemicystectomy versus Segmental Resection: Based on recenthuman and dog experiments,2 the author prefers to perform hemi-cystectomy in place of the previously used segmental resection inwhich the tumor and only a cuff of normal bladder were excised.In practice, hemicystectomy often involves sacrificing 1/3 to 2/3 ofthe organ with reimplantation of a ureter into the bladder. Prior toreconstruction of the bladder, lymphadenectomy is performed. Wherethe tumor is within 2 to 4 centimeters of the bladder neck, moreradical surgery is employed (cystectomy). In addition, hemicystectomyis reserved for those patients in whom the tumor had been proven bybiopsy to invade only the superficial layers of the muscle or less.

While it is true that hemicystectomy is probably unnecessary forvery small bladder cancers, it is impossible to determine this fact selec-tively preoperatively at present without studying a larger series ofpatients treated by this method (sixteen at this writing). Experimentaland clinical data obtained to date indicate that many of these non-

9 2 4 THE BULLETIN

925

infiltrative cancers have spread a great distance within the bladder wallaway from the small lesion grossly visible within the bladder. Regardlessof whether hemicystectomy or segmental resection is performed, itprobably should be in conjunction with lymphadenectomy.

Cystectomy: Cystectomy should not be reserved for the far-ad-vanced cancer but rather for the tumor proven by biopsy not to havespread through the entire bladder wall. The use of cystectomy for infil-trative cancer is a surgical attempt to cure the surgically incurable inthe opinion of the author. Previously we performed this operation forfar-advanced cases with the same poor results reported by others. Laterwe utilized cystectomy for all early non-infiltrative vesical cancers.During the past several years it has been our opinion that cystectomymay not be necessary to effect cure of bladder cancers that are notwithin two to four centimeters of the bladder neck. Cystectomy requiresabout one hour of operating time, but the lymphadenectomy takesthree to four hours. It should be stressed that lymphadenectomy is notto be construed as simple removal of lymph nodes, but clean, completestripping of all tissues off of the internal, external, and common iliacarteries and veins, and also the lower aorta and vena cava.

Radiation Therapy for Non-Infiltrative Vesical Cancer: Radiationtherapy has a definite though incompletely defined place in treatment ofbladder cancer. Certain factors have been established and should be dis-cussed. All radiation therapy is based on tumor dose. This dose may bedelivered by external radiation or internal radiation by radium bomb,radon seed, radioactive silver wire or radioactive isotopes. As withexcision of non-infiltrating cancers, the results of radiation therapydepend also on depth of infiltration of the cancer within the bladderwall. Statistics (particularly those from England and Sweden) comparefavorably with those obtained by surgical methods of therapy for non-infiltrating vesical tumors. While radiation has the distinct advantageof sparing the patient surgery and its concomitant risks, it has associateddisadvantages. Radiation of the. bladder produces severe vesical spasmand irritability. Not infrequently ureteral strictures and hydronephrosisdevelop. Radiation therapy may or may not kill a localized vesicalcancer (it is impossible to determine without subsequent surgical biopsyof the lesion), but it does not treat lymphatic spread. Perhaps treatmentof non-infiltrative bladder cancers may be managed best by preoperativetumor radiation followed in two or three months by lymphadenectomy.

Bladder Cancer 9 2 5

92 H BLEI

In management of a similarly acting tumor, Taussig and later Meigs,have obtained excellent results in treatment of early cervical cancers bythis attack. As all bladder cancers are generally quite resistant to radia-tion therapy, a combination of radiation and surgery or surgery aloneshould probably effect the greatest cure rate.

COMPLICATIONS AFTER DIVERSION OF THE URETERS

With more than sixty techniques for uretero-intestinal anastomosisalready described in the literature, it would appear that the cause of thefrequent unsatisfactory-results may not depend per se on the methodused but on other more fundamental factors. It has long been consideredthat stricture formation secondary to the normal healing process at thesite of the anastomosis has been responsible for the hydroureter, hydro-nephrosis, pyelonephritis and uremia so often observed. A method tominimize this inherent defect by direct mucosal anastomosis has beenproposed by Nesbit and others. It was the objective of a previouslyreported investigation from this laboratory3 to point out that periureteralfibrosis and the resulting stricture was not the sole contributor to thisdistressing problem. Ureterorenal reflux of fecal-contaminated urine andimpairment of conduction of peristalsis following ureteral implantationappeared to warrant consideration. Data obtained from these threestudies will be discussed.

Ureterorenal Reflux: Reflux may occur when two conditions obtain:First, a patent ureteral stoma must be present; and second, intracolonicpressure must exceed intraureteral pressure. In the patients studied itwas shown that after ureterosigmoid anastomosis the pressure withinthe colon exceeds that within the ureter before the defecation reflexis initiated thereby producing reflux. From our observations should thepatient experience the desire to "void" he probably already has uretero-renal reflux of fecal-contaminated urine. The value of frequent "void-ings" is obvious. Rectograms demonstrating this phenomenon were ofresting pressures which does not take under consideration the knownfact that intracolonic pressure is elevated ioo per cent during evacuation.Were this latter aspect investigated it would unquestionably revealreflux of a greater degree than the 8 to i 5 per cent incidence in ourseries and that of others.

An additional factor must necessarily be considered. Following anyof the accepted techniques of uretero-intestinal implantation a certain

9 2 6 THE BULLETIN

927

percentage of patients will have normal upper urinary tracts. Graveshas noted that most patients comprising this category have nonmalignantdisease. Our experience has been nearly identical, but may be extendedto include all patients having normal ureters on preoperative pyelo-grams. It was interesting to observe that preoperative pyelograms werenormal in all patients having normal postoperative films regardless ofdiagnosis. In these patients the crux of the explanation may dependlargely on the preoperative physiological status of the ureter. This con-cept is not new. Draper and Braasch in I9I134 Cut the ureterovesical valvein normal dogs and found that reflux did not occur in most instances.They concluded that the check-valve action of the intravesical ureter isof contributory rather than paramount importance. In the absence ofthe valve, reflux may be prevented by the normal physiologic functionof the ureter which compensates for the loss. This impression has beenconfirmed by our clinical data. In our hands the best results have beenobtained by uretero-intestinal anastomosis if preoperative pyelogramsrevealed ureters of normal caliber, and by ureterocutaneous anastomosisif the x-rays demonstrated ureteral dilatation. Two explanations of thisare possible: First, unless a deliberate narrowing of the ureteral stoma isproduced (which then acts as a stricture), due to its greater circumfer-ence the dilated ureter anastomosed to the bowel is usually more subjectto ureterorenal reflux than the normal calibered ureter; second, ureteraldilatation indicates increased retrograde pressure and frequently asso-ciated renal tubular disease. In patients with renal tubular damage minordegrees of acidosis may assume alarming proportions. Ureterocutaneousanastomosis precludes the possibility of hyperchloremic acidosis fromurinary rectal dialysis observed in some patients with uretero-intestinalanastomosis.

Where uretero-intestinal anastomosis is employed, it has been deter-mined previously5 in humans and dogs that implantation of the uretersinto the mid-descending colon creates a large rectosigmoid reservoir forurine below the stomata and reduces considerably the incidence of re-flux. The level of urine within the rectosigmoid does not reach theheight of the ureteral stomata prior to initiation of the defecating reflexunless the patient is recumbent. This important advantage is not an un-mixed blessing, however, as it produces a slightly higher incidence ofacidosis unless the patient is advised carefully relative to the necessityof frequent "voidings."

Bladder Cancer 9 2 7

9l H BLEI

Ureteral Action Potentials: It is conceivable that the extremely highincidence of ureteral dilatation subsequent to uretero-intestinal anas-tomosis might be related to impairment of conduction of peristaltic actionpotentials following division and implantation of the ureter. In a reportfrom this laboratory," a study has been presented of an experimentalinvestigation of ureteral action potentials in the normal dog. Theseresults were then compared with electro-ureterograms obtained afterre-anastomosis of the divided ureter to bladder, uretero-intestinal im-plantation and incomplete and complete occlusion of the ureter. In brief,the findings of this project demonstrated that the normal electro-uretero-gram is usually triphasic in form. An occasional normal variant may bediphasic. Normal peristalsis is rarely associated with after-potentials.Transection of the ureter below the electrode does not significantlyalter the normal electroureterogram. Re-anastomosis of the transectedureter to the bladder or implantation in the bowel also failed to producesignificant abnormalities of peristaltic action potentials unless ureteraldilatation developed as a result of stricture. In control experiments theseabnormalities were demonstrated also in the dilated, but intact, ureter,indicating these changes to be associated with ureteral dilatation and un-related to impairment of propagation of conduction in a previouslytransected ureter. To a limited degree this study has been followed alsoin humans with identical results. It may be stated with some measureof confidence, therefore, that ureteral dilatation subsequent to uretero-intestinal anastomosis is in no way related to impairment of conductionof peristaltic action potentials.

Uretero-Intestinal Stricture: Probably the most important contribu-tion in this connection was the investigation of Hinman and Weyrauchin I942.7 Their study consisted of uretero-intestinal and ureterovesicalimplantation in dogs. It was observed that healing at the stomal siteoccurred either by primary union and early epithelialization (re-anas-tomosis of ureter to bladder), or by granulation and cicatrization andlate epithelialization (uretero-intestinal implantation). The latter condi-tion obtained in the presence of infection and the dogs developed stric-ture, hydroureter, hydronephrosis and pyelonephritis.

This experimental demonstration of periureteral concentric fibrosisand stricture at the area where ureter traverses bowel wall has greatclinical significance as a large percentage of patients with uretero-in-testinal anastomosis have ureteral dilatation and hydronephrosis.

9 2 8 THE BULLETIN

-~~~ ~lde Cacr

Ragan and his associates, Spain and Molomut and others demon-strated that by systemic use of cortisone, connective tissue formationcould be inhibited in experimentally produced wounds. Integration ofthese concepts resulted in experiments reported previously from thislaboratory in which it may be concluded that parenteral administrationof cortisone does not reduce the normal amount of fibrous tissue in anuninfected healing wound uniting ureter to bowel. Infection at oraround the site of anastomosis, however, acts as a stimulus for increasedlocal deposition of fibrous tissue with resultant stricture formation. Ap-parently cortisone modifies or reduces this fibroplasia or stimulus-pro-ducing fibroplasia in the infected healing wound. This observation wasconfirmed by additional data published on the prevention and treatmentof other ureteral and urethral strictures of the urological tract.8 It is be-lieved that this beneficial effect of scar tissue inhibition in the healingwound by use of cortisone is observed only in the presence of infection.It would appear to be of importance, therefore, to reduce to a minimum,bowel infection during the reparative stage by administering intestinalantibiotic drugs until complete healing of ureter to bowel has beeneffected. It has been determined previously that although cortisoneretards fibroplasia in wounds, a firm union is effected in a maximum ofthree weeks. Cortisone and intestinal antibiotics may probably be safelydiscontinued at that time without subsequently developing a uretero-intestinal stricture. Recent unreported observations indicate that the useof cortisone is probably of little value if sterilization of the bowel ismaintained for three weeks postoperatively.

Acidosis and Pyelonephritis: The development of hyperchloremicacidosis remains as one of the principle complications after uretero-in-testinal anastomosis. The responsible factor is dialysis of urinary chlorideacross the rectal mucosa before the defecation reflex is initiated. As thiscomplication cannot be eliminated completely with this form of diver-sion of the ureters, these patients require indefinite follow-up in theoffice or out-patient department. Periodically, serum chloride and car-bon dioxide combining power must be determined. This conditionshould be suspected if the patient develops weakness, leg pains or an-orexia, or fatigues easily. Acidosis, if present, is usually easily controlledwith oral administration of sodium lactate or bicarbonate and urgingthe patient to "void" more frequently. Acidosis is usually greatly exag-gerated- in the presence of hydronephrosis. There is still some con-

Bladder Cancer 9 2 9

troversy as to whether acidosis can ever develop in the absence of renaldisease. It is the opinion of the author that patients with moderately tomarkedly dilated ureters should have ureterocutaneous rather thanuretero-intestinal anastomosis to nearly eliminate this complication.Ureterocutaneous anastomoses tend to develop ureteral retraction orstrictures which must be dilated periodically.

Subsequent to diversion of the ureters (to skin or intestine), theremay be occasional bouts of pyelonephritis. Most of these are adequatelycontrolled with Gantrisin, but persistence of temperature or chills, back-ache, rectal tenesmus -or even incontinence, indicates the need of adifferent antibiotic agent. Treatment should be vigorous and continuedfor at least one week after complete subsidence of signs and symptoms.Defunctioning the lower bowel of feces by a preliminary and perma-nent colostomy, after the method of Vest and Boyce, has a pronouncedeffect in preventing these infections.

DISCUSSION AND CONCLUSIONS

In aniline dye workers, clinical and laboratory investigations havedemonstrated at least one etiological factor in development of bladdercancer. It has been shown that depth of infiltration of the cancer in thebladder wall is highly significant with regard to both therapy andprognosis. Potentially, non-infiltrative cancers that are treated by ac-cepted, present-day techniques should result in about an 85 per centcure. Infiltrative tumors treated by any or all methods should have acure rate of about IS per cent or less. It would appear from investiga-tions in humans and dogs that if ureteral diversion is to be employed,patients will obtain the best results if dilated ureters are transplanted tothe skin and non-dilated ureters are implanted in the bowel. Treatmentof infiltrative bladder cancer is directed at palliation by transurethralresection, and in selected cases, diversion of the urinary stream. Statisticsto date have demonstrated that cystectomy is usually not curative inthese patients. In treatment of non-infiltrative bladder tumors the objec-tions to transurethral resection have been indicated. Cystectomy or gen-erous segmental bladder resection appear to be more reasonable thera-peutic procedures.

Studies of vesical lymphatics in dogs and spread of non-invasivebladder tumors in humans have indicated that conventional segmentalresection may not be satisfactory treatment. Hemicystectomy in con-

THE BULLETIN9 3 0

-~~~~~lde Cace 93

junction with lymphadenectomy appears more reasonable from data nowavailable. What is visually only a small bladder tumor may have spreadmicroscopically throughout one-third of the bladder wall. About 2 to 5per cent of non-infiltrative cancers have lymphatic spread. Cystectomyor hemicystectomy without lymphadenectomy appears to be an inade-quate procedure.

Whether or not the lower colon should be defunctionalized by per-manent colostomy after uretero-intestinal anastomosis is still underlaboratory and clinical investigation. Transection and implantation ofthe ureters to skin or bowel do not produce alteration of peristalticaction potentials. Experiments in dogs demonstrated that ureteral dilata-tion following uretero-intestinal anastomosis is usually the result of peri-ureteral concentric fibrosis where the ureter heals to the bowel wallwith the production of a stricture. This may be reduced to a large degreeafter ureteral implantation by the postoperative administration of in-testinal antibiotic drugs to hold the flora of the bowel at a minimum untilthe ureter has healed completely to the bowel. Experimental investigationhas indicated that about 8 to i 5 per cent of patients with ureteral dilata-tion after transplantation have ureterorenal reflux of fecal-contaminatedurine. This may be reduced by anastomosing the ureters to a slightlyhigher level in the bowel and having the patients "void" more fre-quently. The problem of. hyperchloremic acidosis following uretero-intestinal anastomosis has been studied in humans and dogs and the con-clusion is that it is due to selective reabsorption of chloride within therectum prior to "voiding" and is aggravated in the presence of renaltubular damage. Adequate rectal evacuation and administration of alkaliwhen indicated control this complication. Pyelonephritis may developafter ureteral diversion to skin or bowel and regular follow-up visitsmust be made by these patients.

Previously, most patients developed severe renal disease after uretero-intestinal or ureterocutaneous anastomosis. Clinical and laboratory in-vestigations have contributed toward reducing these complications toa degree where it is observed in only a small percentage of patients sotreated. It is reasonable to expect that further research will solve thisaspect of the problem. In the meanwhile, additional controlled data arebeing collected and reported concerning more specific indications forthe various therapeutic methods available for treatment of cancer of thebladder.

Bladder Cancer 9 3 I

9 3 2 THE BULLETIN

R E FE R E N C E S

1. Jewett, H. J. and Strong, G. H. Infil-trating carcinoma of the bladder: rela-tion of depth of penetration of thebladder wall to incidence of local exten-sion and metastases, J. Urol. 55:366-721946.

2. Baker, R. Hemicystectomy versus seg-mental resection in treatment of non-infiltrating bladder cancer, J. Urol., inpress.

3. Baker, R., Govan, D., Huffer, J. andCason, J. Physiology of the uretero-intestinal anastomosis; inhibitory effectof cortisone on stricture formation, J.Urol. 70: 58-67, 1953.

4. Draper, J. W. and Braasch, W. F.Function of the ureterovesical valve;

an experimental study of the feasibilityof ureteral meatotomy in human beings,J. Amer. med. Assoc. 60:20-24, 1913.

5. Baker, R. and Miller, G. H., Jr. Physi-ology of the uretero-intestinal anasto-mosis; ureteral reflux, J. Urol. 67:638-43, 1952.

6. Baker, R. and Huffer, J. Ureteral elec-tromyography, J. Urol. 70:874-83, 1953.

7. Hinman, F. and Weyrauch, H. M. Ex-perimental study of ureterointestinal im-plantation; destiny of the implanted ure-ter, Sutrg. Gynec. Obstet. 74:129-36, 1942.

8. Baker, R., Govan, D. and Huffer, J.Inhibitory effect of cortisone on stric-tures of the urological tract, Surg.Gynec. Obstet. 95:446-54, 1952.

DATA ON INTERNATIONAL CONGRESSESREQUESTED BY LIBRARY

Fellows are reminded that the Library is eager to acquire the publica-tions of all international congresses, their programs, bulletins, proceed-ings, reports. Since these are often hard to find, it would help a greatdeal if Fellows attending such congresses would turn over to the Libraryany such papers as they themselves do not wish to keep. Such thought-fulness would be deeply appreciated.