当前位置: 首页 > 期刊 > 《新英格兰医药杂志》 > 2004年第20期 > 正文
编号:11306563
Case 34-2004 — A 45-Year-Old Man with Recurrent Pain in the Right Flank and Hematuria
http://www.100md.com 《新英格兰医药杂志》
     Presentation of Case

    Dr. Donald S. Kaufman (Hematology–Oncology): A 45-year-old man was admitted to the hospital because of recurrent flank pain, hematuria, and a mass in the renal pelvis.

    The patient had had an episode of hematuria and pain in the right flank 11 months earlier, which had been evaluated at another hospital. At that time, an intravenous pyelogram showed a high-grade obstruction on the right side. Right retrograde nephroureterography performed at the same time a stent was placed showed a filling defect in the right proximal ureter that was consistent with a uric acid stone. A radiograph after the procedure showed the stent in good position and also showed at least two areas of density indicating calcification that were believed to be calculi; both were 1 cm in diameter and were in the region of the middle of the right kidney.

    Two months later, the stent was removed because of associated pain and an increase in both urinary urgency and frequency. A computed tomographic (CT) scan of the abdomen and pelvis showed a calcific opacity, 2 mm in diameter, within the lower pole of the right kidney that was believed to be a nonobstructing renal stone. The patient had been well otherwise. He drank alcohol rarely and had smoked cigarettes in the past, but had stopped 15 years previously.

    Two weeks before the admission to this hospital, he had had a sudden, sharp pain in the right flank that radiated to the right groin and was associated with mild hematuria. He was evaluated five hours later in an emergency department near his home. The temperature was 36.9°C, the blood pressure 142/76 mm Hg, the pulse 96 beats per minute, and the respiratory rate 24 breaths per minute. An intravenous pyelogram showed a dilated collecting system on the right and no contrast material in the right ureter. A diagnosis of a stone in the right kidney was made; hydromorphone and ketorolac tromethamine were administered for pain, and promethazine for nausea. Urinalysis showed a pH of 7.5; the urine was positive for occult blood (+++), the red cells were too numerous to count, and no white cells or bacteria were seen. The patient's symptoms improved, but during the next seven hours the pain returned, with epigastric discomfort, chills, and a temperature as high as 38.5°C. He was given additional analgesic and antiemetic medication, levofloxacin, and famotidine. He was transferred to this hospital.

    On examination, the temperature was 36.6°C, the blood pressure 151/100 mm Hg, the pulse 70 beats per minute, and the respiratory rate 20 breaths per minute. The abdomen was soft, with mild tenderness in the right lower quadrant and right flank; there were no masses. The examination was otherwise normal. Urinalysis showed a pH of 6.5, and the urine tested positive (+++) for occult blood; there were between 50 and 100 red cells, between 5 and 10 white cells, and a few bacteria per high-power field. A complete blood count; electrolyte levels; results of liver-function tests; the values for uric acid, urea nitrogen, creatinine, protein, amylase, and lipase; and the prothrombin and partial-thromboplastin times were all within normal ranges. Meperidine, hydroxyzine, morphine sulfate, ampicillin, and gentamicin were given, and fluids were administered intravenously. The fever and pain resolved over the course of the following nine hours.

    On the next day, a CT scan of the abdomen without the administration of contrast material showed a nonobstructing stone, 3 mm in diameter, in the lower pole of the right kidney (Figure 1A). There was mild dilatation of the right renal collecting system, with periureteral and perirenal stranding suggestive of a recently passed stone (Figure 1B). Ultrasonographic examination of the right kidney showed a soft-tissue density within the right renal pelvis with areas of high attenuation that were believed to represent blood clots (Figure 2). There was no evidence of hydronephrosis or masses within the kidneys. No calculi were identified. The patient was discharged later that day with instructions to strain his urine for stones, take levofloxacin for two weeks, take oxycodone and acetaminophen for pain, and return if worsening hematuria, severe pain, or fever developed.

    Figure 1. Computed Tomographic (CT) Scans of the Abdomen and Pelvis.

    CT scanning without the administration of contrast material (Panel A) showed a nonobstructing stone, 2 to 3 mm in diameter (arrow), in the lower pole of the right kidney in a location where calcific densities had been seen several months earlier. In contrast to what was shown on the earlier CT scan, there was mild dilatation of the right collecting system (Panel B) that contained high-density material (arrows) within the pelvocaliceal system, consistent with the presence of blood. CT scanning after the administration of contrast material 11 days later (Panel C) shows an irregular filling defect (arrow) within the right pelvocaliceal system, with no other abnormalities.

    Figure 2. Ultrasonogram of the Kidney.

    An area of soft-tissue density is visible within the pelvocaliceal system.

    Eleven days after discharge, he returned as an outpatient for follow-up CT of the abdomen after the oral and intravenous administration of contrast material. The scan showed a nonobstructing stone, 0.4 cm in diameter, in the lower pole of the right kidney, with a density of 300 Hounsfield units. Delayed images, which show the excretory phase better, were obtained after the administration of contrast material, and they revealed an irregularly shaped soft-tissue mass, 2.6 cm by 2.0 cm, in the right renal pelvis (Figure 1C).

    Three days later, the patient was readmitted to the hospital, and a diagnostic procedure was performed.

    Differential Diagnosis

    Dr. Kaufman: May we review the radiology studies?

    Dr. Peter R. Mueller: The films from the other hospital are not available.

    At the time of the man's first admission here, a routine "stone protocol" CT study was performed; this consists of a helical CT scan of the abdomen and pelvis obtained without the administration of either oral or intravenously administered contrast material.1 This is now considered the standard approach to the evaluation of suspected stone disease. The resulting scans showed a nonobstructing stone, 2 to 3 mm in diameter, in the lower pole of the right kidney in a location where calcific densities had been seen several months earlier (Figure 1A). However, on this CT scan there was mild dilatation of the right collecting system that contained high-density material within the pelvocaliceal system, consistent with the presence of blood (Figure 1B). There was perirenal and periureteral stranding, but no stone was seen in the pelvocaliceal system or ureter. No specific cause of hydronephrosis was noted.

    This study was followed by an ultrasonographic examination on the same day to evaluate the collecting system. The ultrasonogram showed a soft-tissue density within the pelvocaliceal system with associated hydronephrosis (Figure 2). It was difficult to determine on this study whether the dense area was a clot or a tumor.

    Eleven days later a specialized CT procedure to evaluate hematuria was performed. This consists of standard CT that extends through the kidneys, ureters, and bladder, which is then followed by CT scanning performed after the intravenous administration of contrast material, with delayed images. This delay allows the pelvocaliceal system, ureter, and bladder to be imaged in the excretory phase.2 Tumors of the kidney, parenchyma, and pelvocaliceal system can be evaluated by this method. Delayed imaging makes possible visualization of the collecting system, ureters, and bladder; the result is analogous to an intravenous pyelogram, but with better resolution. On this examination, an irregular filling defect was noted within the right pelvocaliceal system, with no other abnormalities (Figure 1C). When the density of this filling defect on images obtained with and without added contrast material was compared, evidence of enhancement with contrast material was seen.

    Dr. Kaufman: Dr. Mueller, does a case like this require radiologists to rethink the standard approach to hematuria, especially when it occurs in a younger person?

    Dr. Mueller: The role of radiologic imaging in patients with suspected stone disease or hematuria, or both, has changed dramatically in the past five years.3 CT, because of its ability to scan thin sections in short periods of time and because of the availability of multiplanar reconstruction, has replaced intravenous urography as the primary diagnostic tool in both settings.1,2 A CT scan obtained without the administration of contrast material is used to diagnose a stone; the addition of intravenously administered contrast material with delayed images is used to diagnose tumors. The accuracy and sensitivity of a non–contrast-enhanced CT scan for renal, ureteral, and bladder stones approach 100 percent.4 Tumors as small as 1 cm in diameter can be visualized by contrast-enhanced CT scanning.5

    Dr. Kaufman: Dr. Harisinghani, should every patient with hematuria undergo a contrast-enhanced study?

    Dr. Mukesh Harisinghani (Radiology): Yes. Until recently, the imaging algorithm for evaluating patients with hematuria was to obtain a scan without enhancement and if stones were identified, no contrast material was administered. However, small tumors can be missed with this technique. With the use of intravenous contrast material, such tumors can be diagnosed.

    Dr. Kaufman: I have asked Dr. Dahl, who cared for this patient, to review the differential diagnosis of hematuria and discuss the evaluation and management of this case.

    Dr. Douglas M. Dahl: This 45-year-old man, with a history of cigarette smoking, had recurrent episodes of right renal colic and hematuria and radiographic findings of renal obstruction and a stone.

    The differential diagnosis of hematuria is broad and includes pathologic processes in the renal parenchyma, upper urinary collecting system, ureter, bladder, and urethra, as well as extrinsic processes that encroach on the urinary drainage system.6 The focus is narrowed by the knowledge of a localizing symptom and radiographic findings of an obstruction in the upper urinary tract. I will discuss sources of bleeding from the upper urinary tract that could explain this patient's clinical syndrome.

    Urinary stone disease is common, occurring in 1 in 10 people over a lifetime. Symptoms include severe colicky flank pain. Gross hematuria is present in 15 percent.3 The renal pelvis and ureter contain highly vascular epithelium that, when irritated, can cause bleeding. Blood clots may form and obstruct the flow of urine, which results in pain. Stones may contain either calcium or uric acid. A patient with uric acid lithiasis typically has a very low urinary pH value. Uric acid is insoluble at a pH of 5.75 but soluble above this range.7 In this patient the urinary pH was 6.5. Primary neoplasms of the renal parenchyma or transitional-cell epithelium must also be considered. The most common renal neoplasm is renal-cell carcinoma. Neoplasms may cause microscopic or gross hematuria. Either the tumor or the blood clots can obstruct the collecting system, which results in the clinical picture of pain.

    The overwhelming likelihood at the time of his initial presentation was that this patient's hematuria was the result of a stone or another benign process. At the time his stent was placed, the urologist would have evaluated the urethra and bladder for other findings that could explain hematuria. The part of the picture that does not fit with the expectations is that the patient had a tiny stone that did not appear to be in a location to cause obstruction, pain, or gross hematuria. At the time of his second presentation, with a large amount of material that could be blood in the renal pelvis, the stone had not moved. This finding raises the concern that the material in the pelvis may represent another pathologic process.

    Ultrasonography should easily distinguish a stone from a clot or a tissue mass. Tumors tend to be adherent to the pelvis and have an irregular contour, whereas stones are not adherent and have a smooth contour. In this case, ultrasonography showed that the defect was not a stone, but the procedure could not distinguish between a clot and a tumor. The results of cytologic examination of the urine are abnormal in most cases of high-grade transitional-cell tumors. Unfortunately, cytologic samples are less commonly diagnostic for low-grade neoplasms.3

    The finding of a filling defect in the renal pelvis makes the possibility of a primary neoplasm the immediate focus in the clinical evaluation. Transitional-cell tumors of the renal pelvis account for 95 percent of neoplasms arising in the renal collecting system.8 Since these tumors can be highly aggressive, an evaluation of suspected transitional-cell tumors should be done as soon as possible. Transitional-cell carcinomas have been estimated to make up 5 percent of all renal masses,8 but in my experience this is an overestimate. This disease appears most often in the sixth and seventh decades of life. It is more common in men than in women. Cigarette smoking increases the risk of upper urinary tract transitional-cell tumors by a factor of as much as 4.5.8 This man with a history of smoking thus has some risk factors for this disease.

    In this patient's case, the clinical diagnosis was a primary transitional-cell carcinoma of the renal pelvis. The next step in the evaluation was endoscopic exploration of the bladder and ureter under general anesthesia. Cystourethroscopy with a careful examination of the bladder was performed, since pelvic tumors are often associated with other lesions, most commonly in the bladder. Cystoscopy revealed a small erythematous area on the posterior right side of the bladder wall, and a biopsy specimen was obtained.

    Retrograde urography, in which standard intravenously administered radiographic contrast material is passed by way of a catheter up the ureter and into the renal pelvis, was then performed. This test is highly sensitive for detecting dilatation of the collecting system and space-occupying lesions, which appear as filling defects. Once the catheter is placed in the ureter, but before the instillation of contrast material, samples can be obtained for cytologic examination. In this case, retrograde pyelography demonstrated a filling defect in the renal pelvis, which corresponded to that seen on the CT scan.

    Unless active, brisk bleeding limits visualization, ureteroscopic evaluation is helpful in further characterizing the source of bleeding in the upper urinary tract. The use of flexible ureteroscopes permits visualization of the ureter, renal pelvis, and all major renal calices. Most stones can be identified, broken up, and removed. Primary neoplasms can be seen and biopsy specimens can be obtained. Percutaneous access to the kidney, followed by endoscopic evaluation of the renal pelvis, is possible, but this technique is more invasive and can increase the risk of bleeding and tumor dissemination.9,10 In this case, ureteroscopy showed a large pedunculated papillary neoplasm of the renal pelvis, with no other abnormalities. Biopsy specimens and washings were obtained for cytologic evaluation, and a stent was placed in the renal pelvis.

    Dr. Douglas M. Dahl's Diagnosis

    Transitional-cell carcinoma of the right renal pelvis.

    Pathological Discussion

    Dr. Robert H. Young: The small fragment of tissue that was examined microscopically showed the typical features of a histologic grade 1 papillary transitional-cell carcinoma of the urinary tract (Figure 3A). Delicate fibrovascular cores were covered by hyperplastic, minimally atypical transitional cells. There was no evidence of invasion, but the limited amount of tissue available for evaluation left open the possibility of invasion in unsampled tissue. The specimen from the bladder showed no abnormalities. Cytopathological examination of the fluid from the renal pelvis showed atypical urothelial cells in clusters, which may suggest low-grade papillary transitional-cell carcinoma.

    Figure 3. Histologic Sections of the Mass in the Renal Pelvis (Hematoxylin and Eosin).

    The biopsy specimen (Panel A) represents a typical low-grade papillary transitional-cell carcinoma, in which delicate fibrovascular cores are covered by hyperplastic, minimally atypical transitional cells. The specimen from the resected tumor (Panel B) has an unusual inverted pattern of growth, with large nests of tumor cells within the lamina propria. This does not represent true invasion. At higher magnification (Panel C), the tumor cells resemble normal transitional cells, as is typical of a low-grade tumor.

    The spectrum of epithelial neoplasms derived from the renal pelvic epithelium parallels that in the urinary bladder and includes transitional-cell carcinoma and its variants, rare occurrences of pure squamous-cell carcinoma and pure adenocarcinoma, and even benign neoplasms, such as the inverted papilloma. As in the urinary bladder, the transitional-cell carcinomas range from being low grade and noninvasive with a good prognosis to high-grade, deeply invasive tumors, with a poor prognosis.

    Dr. Kaufman: Dr. Dahl, how did you treat this patient?

    Discussion of Management

    Dr. Dahl: This patient had an uncommon tumor, a transitional-cell carcinoma of the renal pelvis. However, the physicians at this hospital who reviewed the images agreed that it was localized, without metastatic disease. The biopsy specimen showed a low-grade tumor, and tumors such as these are likely to be noninvasive. The patient was otherwise healthy.

    The standard management of a primary neoplasm of the upper urinary tract is radical nephroureterectomy, with removal of the kidney and entire ureter. Transitional-cell carcinomas may be multifocal, and cytologic atypia is frequently found in areas of grossly uninvolved renal pelvis and ureter. In cases in which a primary lesion of the renal pelvis is treated only by nephrectomy, at least 16 percent of patients will later have neoplasms within the ipsilateral ureter.11,12 Before a decision could be made about surgical management in this particular case, we had to evaluate the patient's overall fitness for major surgical intervention, including an evaluation of renal function, to predict whether there would be detrimental consequences from the removal of one kidney.

    Small tumors of the renal pelvis can be removed by noninvasive techniques. One approach is ablation by ureteroscopy, with the use of laser or electrical energy. The holmium laser, which is commonly used to treat kidney and ureteral stones, can be used for the ablation of certain tissue lesions. Ureteroscopy can be performed in an outpatient surgical setting. However, several studies show that tumors may be assigned a lower stage when evaluated by endoscopic biopsy and then turn out to be more aggressive and invasive than suspected.13 In addition, several procedures may be required to accomplish successful ablation, and vigilant surveillance for recurrence must be conducted; thus, the patient is committed to undergoing many procedures under general anesthesia.

    Another approach would be removal by means of nephroscopy.9 After the placement of a nephrostomy tube into the renal calyx, nephroscopy is commonly performed to remove large renal stones, and it can be used to remove tumors. Standard transurethral resectoscopic instruments can be used to resect the tumor and ensure adequate hemostasis. Adjacent tissue can be biopsied to rule out cytologic atypia. With an increase in the manipulation of the renal parenchyma, however, there is a substantial increase in the possibility of damage to the major branches of the renal vasculature. It is also important to be aware of the possibility of the implantation of tumor cells outside the renal collecting system, into the retroperitoneum or along the percutaneous access tract.

    Because of their limitations and complications, I believe that ureteroscopic and percutaneous resection should be reserved for those rare circumstances in which the patient cannot tolerate radical nephrectomy, either because of poor renal function or the presence of a solitary kidney, and when there is confirmation that the tumor is of a low grade and superficial.

    In a young, healthy patient with normal renal function and two kidneys, complete surgical removal of the kidney and ureter, including the intravesical portion, is indicated. In addition, lymph-node dissection should be performed along the right pericaval region. This procedure can be accomplished by an open surgical technique, through either two incisions — a flank incision to mobilize and resect the kidney followed by a lower abdominal incision to resect the distal ureter and its intravesical segment — or through one large abdominal incision. This highly invasive intervention is the standard of care in many institutions today.

    Advances in optical technology and instrumentation have allowed surgeons to perform nephroureterectomy and retroperitoneal lymph-node dissection with laparoscopic techniques. A complete resection of the kidney, the surrounding renal fat, adjacent lymph-node tissue, and the entire ureter is possible, typically with five small laparoscopic ports. Published case series from several institutions have validated the efficacy of this procedure and demonstrated substantially improved morbidity.14,15 Because of the propensity of transitional-cell carcinoma to spread and seed, particularly within the peritoneal cavity, this procedure should only be performed by experienced laparoscopic surgeons.

    After a discussion of the options, this patient elected to undergo a laparoscopic nephroureterectomy. This procedure was performed two weeks after the biopsy (Figure 4). We removed the kidney and ureter, dissecting the ureter well into the muscle by the transabdominal approach, but we stopped short of entering the mucosa of the bladder. Surgical clips were left at the most distal portion of the ureter, and the ureter was clipped proximally before it was divided to prevent tumor spillage. We also performed a retroperitoneal lymph node dissection along the vena cava. The patient had no complications and was discharged three days later.

    Figure 4. Laparoscopic Technique for Radical Nephroureterectomy.

    In laparoscopic radical nephroureterectomy, the kidney and ureter are mobilized and dissected away from the peritoneum and vessels. Careful circumferential dissection (Panel A) is used to divide the detrusor muscle fibers from the intramuscular portion of the ureter. Traction is placed on the ureter until the entire distal ureter is visualized as it enters the bladder. The distal ureter is ligated with surgical clips (inset). The proximal ureter is ligated, and the ureter is divided. In this patient, five months later at cystoscopy (Panel B), the distal right ureteral tunnel was calibrated and estimated to be 3 mm long. A resectoscope sheath was introduced into the bladder and a knife was used in marsupialization of the ureteral tunnel. The surgical clip left at the time of nephrectomy was extracted with biopsy forceps. The transitional epithelium at the base was coagulated, eliminating any remaining tumor tissue.

    Dr. Young: The Pathology Department received a kidney with an attached ureter, 27 cm long, and surrounding soft tissue. When the kidney was sectioned, a friable papillary neoplasm, 4 cm at its maximal dimension, was visible within the renal pelvis. Microscopical evaluation showed a papillary pattern similar to that seen in the biopsy specimen. The morphology was somewhat unusual because of the presence of large bulbous aggregates of low-grade neoplastic transitional cells inverting into the underlying lamina propria (Figure 3B and Figure 3C). The picture was similar to that seen in some transitional-cell carcinomas of the urinary bladder, and is referred to as an inverted pattern of growth.16 This picture is different from that seen in a benign urinary tract neoplasm, the inverted papilloma, which has a more compact arrangement and generally lacks a surface papillary component. The inverted growth pattern is not considered to represent true invasion, which produces irregular stromal penetration. There were no lymph-node metastases.

    Dr. W. Scott McDougal (Urology): Dr. Dahl, under what circumstances might you recommend an open nephroureterectomy?

    Dr. Dahl: The most important factor is probably the experience of the surgeon. Patient factors that make laparoscopic surgery more difficult include a history of abdominal surgery, any major procedures involving the kidney, or ureteral surgery. For an invasive tumor with bulky lymph-node involvement, management by laparoscopic techniques becomes more difficult and the length of time required for the procedure could be more than that for an open surgical technique.

    Five months after the nephrectomy, I performed cystourethroscopy while the patient was under anesthesia and performed a marsupialization procedure with the remaining 3 mm of the distal ureter, so that it can now be observed by standard office cystoscopy, and removed the clip (Figure 4B). Surveillance cystoscopy is necessary for any patient with a history of urothelial carcinoma, because of the possibility that other neoplasms will develop, either within the remaining ureter or in the bladder. As a rule, I perform cystoscopy in the office every three months for the first two years after a resection, every six months for five years, and once a year thereafter.17 In this case, 18 months after the nephroureterectomy, the patient continues to undergo regular cystourethroscopic examinations with cytologic examination of urine specimens, and he remains free of disease.

    Anatomical Diagnosis

    Papillary transitional-cell carcinoma of the renal pelvis, low grade.

    Source Information

    From the Departments of Urology (D.M.D.), Radiology (P.R.M.), and Pathology (R.H.Y.), Massachusetts General Hospital; and the Departments of Surgery (D.M.D.), Radiology (P.R.M.), and Pathology (R.H.Y.), Harvard Medical School.

    References

    Fielding JR, Silverman SG, Samuel S, Zou KH, Loughlin KR. Unenhanced helical CT of ureteral stones: a replacement for excretory urography in planning treatment. AJR Am J Roentgenol 1998;171:1051-1053.

    McNicholas MM, Raptopoulos VD, Schwartz RK, et al. Excretory phase CT urography for opacification of the urinary collecting system. AJR Am J Roentgenol 1998;170:1261-1267.

    Grossfeld GD, Litwin MS, Wolf JS Jr, et al. Evaluation of asymptomatic microscopic hematuria in adults: the American Urological Association best practice policy. II. Patient evaluation, cytology, voided markers, imaging, cystoscopy, nephrology evaluation, and follow-up. Urology 2001;57:604-610.

    Schreyer HH, Uggowitzer MM, Ruppert-Kohlmayr A. Helical CT of the urinary organs. Eur Radiol 2002;12:575-591.

    Jamis-Dow CA, Choyke PL, Jennings SB, Linehan WM, Thakore KN, Walther MM. Small (< or = 3-cm) renal masses: detection with CT versus US and pathologic correlation. Radiology 1996;198:785-788.

    Cohen RA, Brown RS. Microscopic hematuria. N Engl J Med 2003;348:2330-2338.

    Gutman AB, Yu TF. Uric acid nephrolithiasis. Am J Med 1968;45:756-779.

    Donat SM, Herr HW. Transitional cell carcinoma of the renal pelvis and ureters: diagnosis, staging, management, and prognosis. In: Oesterling JE, Richie JP, eds. Urologic oncology. Philadelphia: W.B. Saunders, 1997:215-34.

    Goel MC, Mahendra V, Roberts JG. Percutaneous management of renal pelvic urothelial tumors: long-term followup. J Urol 2003;169:925-929.

    Ong AM, Bhayani SB, Pavlovich CP. Trocar site recurrence after laparoscopic nephroureterectomy. J Urol 2003;170:1301-1301.

    Hisataki T, Miyao N, Masumori N, et al. Risk factors for the development of bladder cancer after upper tract urothelial cancer. Urology 2000;55:663-667.

    Nocks BN, Heney NM, Daly JJ, Perrone TA, Griffin PP, Prout GR Jr. Transitional cell carcinoma of renal pelvis. Urology 1982;19:472-477.

    Daneshmand S, Quek ML, Huffman JL. Endoscopic management of upper urinary tract transitional cell carcinoma: long-term experience. Cancer 2003;98:55-60.

    El Fattouh HA, Rassweiler JJ, Schulze M, et al. Laparoscopic radical nephroureterectomy: results of an international multicenter study. Eur Urol 2002;42:447-452.

    McNeill SA, Chrisofos M, Tolley DA. The long-term outcome after laparoscopic nephroureterectomy: a comparison with open nephroureterectomy. BJU Int 2000;86:619-623.

    Amin MB, Gomez JA, Young RH. Urothelial transitional cell carcinoma with endophytic growth patterns: a discussion of patterns of invasion and problems associated with assessment of invasion in 18 cases. Am J Surg Pathol 1997;21:1057-1068.

    Linehan WM, Zbar B, Bates SE, Zelefsky MJ, Yang JC. Cancer of the kidney and ureter. In: De Vita VT Jr, Hellman S, Rosenberg SA, eds. Cancer: principles & practice of oncology. 6th ed. Philadelphia: Lippincott Williams & Wilkins, 2001:1362-96.(Douglas M. Dahl, M.D., Pe)