- Email: [email protected]
Pathologic Findings in Patients with Ureteropelvic Junction Obstruction and Crossing Vessels
Endourology and Stones Pathologic Findings in Patients With Ureteropelvic Junction Obstruction and Crossing Vessels Lee Richstone, Casey A. Seideman, Ernesto Reggio, Rachel Bluebond-Langner, Peter A. Pinto, Bruce Trock, and Louis R. Kavoussi OBJECTIVES
METHODS
RESULTS
CONCLUSIONS
To define the role of crossing vessels in the pathophysiology of ureteropelvic junction (UPJ) obstruction, we analyzed the relationship between the presence of crossing vessels and UPJ pathologic findings in patients undergoing laparoscopic pyeloplasty. The significance of crossing renal vessels in patients with UPJ obstruction is unclear. We performed a retrospective analysis of 155 consecutive patients undergoing laparoscopic pyeloplasty. Pathologic specimens from the UPJ were evaluated in 95 patients. The presence or absence of crossing vessels was documented intraoperatively. The histopathologic findings allowed for categorization into 5 groups: group 1, normal ureteral tissue; group 2, chronic inflammation; group 3, smooth muscle hypertrophy, group 4, fibrosis; and group 5, smooth muscle atrophy. The pathologic findings between patients with and without crossing vessels were compared. Overall, crossing vessels were identified in 98 patients (63.2%). Of the 95 cases with specimens retrieved for histologic analysis, 65 had crossing vessels and 30 did not. The most common UPJ histologic finding in patients with crossing vessels was no intrinsic abnormality (43%). In contrast, this was seen in only 10% of patients without a crossing vessel. In the group without crossing vessels, chronic inflammation (40%) was the predominant histologic findings. Patients with a crossing vessel were less likely to have intrinsic histologic pathologic findings (P ⬍ .0003). Patients with crossing vessels and UPJ obstruction had no histologic abnormalities identified in 43% of cases. This finding implicates crossing vessels in the pathogenesis of select cases of UPJ obstruction and direct mechanical compression as the etiology of obstruction in these individuals. UROLOGY 73: 716 –719, 2009. © 2009 Published by Elsevier Inc.
B
oth intrinsic and extrinsic factors have been associated with ureteropelvic junction (UPJ) obstruction, including intrinsic structural abnormalities, lower pole crossing renal vessels, external mass compression, retroperitoneal inflammation, abdominal aortic aneurysm, calculi, and infection. Obstruction at the UPJ, regardless of the etiology, can result in hydronephrosis, pain, and loss of renal function. The role of crossing renal vessels in patients with UPJ obstruction has been the subject of debate.1-3 The true incidence of lower pole vessels in the general (unobstructed) population is not known. In cadaveric studies, Sampaio1,2 and Sampaio and Favorito4 demonstrated crossing vessels in ⱕ71% of patients dying of various causes; however, it is unclear how accurately this reflects in vivo anatomic relationships. Imaging studies have From the Smith Institute for Urology, North Shore-Long Island Jewish Health System, New Hyde Park, New York Reprint requests: Lee Richstone, M.D., Smith Institute for Urology, North ShoreLong Island Jewish Health System, 450 Lakeville Road, Suite M41, New Hyde Park, NY 11040. E-mail: [email protected] Submitted: April 3, 2008, accepted (with revisions): October 9, 2008
716
© 2009 Published by Elsevier Inc.
suggested an association between crossing vessels and UPJ obstruction. Endoluminal ultrasound studies have demonstrated crossing vessels at the UPJ in 19.2% of patients with a normal UPJ but in 53%-71% of patients with UPJ obstruction.3,5-7 This has been confirmed by additional radiographic studies using conventional angiography, spiral computed tomography angiography, and contrast-enhanced Doppler ultrasonography, documenting crossing vessels in 39%-79% of patients with UPJ obstruction.3 Despite this association, the definitive causal relationship, if any, between crossing vessels and UPJ obstruction remains poorly understood. Numerous studies have attempted to unravel the pathophysiology of UPJ obstruction by histologic evaluation of UPJ tissue.8,9 Although data exist describing the histologic features of the UPJ in patients with UPJ obstruction using both light and electron microscopy,8,9 few if any reports have specifically addressed the histologic findings in patients with and without crossing vessels. Such information has the potential to shed light on the role of crossing vessels in the pathogenesis of UPJ obstruction. We analyzed the histopathologic UPJ specimens of a large series 0090-4295/09/$34.00 doi:10.1016/j.urology.2008.10.069
Table 1. Results of pathologic examination Pathologic Finding
No Crossing Vessel (n ⫽ 30)
Crossing Vessel (n ⫽ 65)
10 (3) 90 (27) 27 (7/27) 40 (11/27) 20 (6/27) 20 (6/27)
43 (28) 57 (37) 9 (3/37) 32 (13/37) 31 (12/37) 31 (12/37)
Normal Abnormal with significant overlap of following histopathologic findings SM hypertrophy Chronic inflammation Fibrosis SM atrophy SM ⫽ smooth muscle. Data presented as percentages, with numbers in parentheses.
of patients with primary UPJ obstruction in an attempt to better understand the mechanisms by which crossing renal vessels might be involved in obstruction.
MATERIAL AND METHODS We performed a retrospective review of 155 consecutive patients who underwent laparoscopic pyeloplasty for primary UPJ obstruction by a single surgeon (L.R.K.). All patients had radiographic evidence of obstruction with symptoms and/or deterioration of renal function. All patients were preoperatively evaluated using computed tomography and renal scanning before surgical intervention. All patients demonstrated moderate to severe hydronephrosis at the evaluation. Additionally, all patients had a renal function of ⱖ20%, with a half-time of ⱖ20 minutes. The mean age was 38 years (range 10-85). No statistically significant correlation was found between the pathologic findings and age. Of the 155 patients, 84 were female and 71 were male. The site of UPJ obstruction was the left in 61 patients (39.4%) and the right in 94 (60.6%). The presence or absence of crossing vessels was documented intraoperatively. Crossing vessels were identified in 98 patients (63.2%). Of those with crossing vessels, 13 underwent nondismembered repair and were excluded from the analysis; all patients for whom pathologic evaluation was done had undergone Anderson-Hynes dismembered pyeloplasty. All patients had stents placed intraoperatively. The patients in whom previous pyeloplasty, antegrade or retrograde endopyelotomy had failed and those with evidence of secondary obstruction were excluded at the discretion of the operating surgeon. This left 95 patients available for analysis in whom the pathologic findings were evaluated, including 65 with crossing vessels and 30 without. Microscopic evaluation was performed with standard hematoxylin-eosin staining techniques. Fischer’s exact t test was used to calculate the P values.
RESULTS Five histologic categories were identified: (a) normal ureteral tissue, (b) chronic inflammation, (c) smooth muscle hypertrophy, (d) fibrosis, and (e) smooth muscle atrophy. The specimens that had abnormal histologic characteristics demonstrated significant overlap in the findings. The results are listed in Table 1. Histologic examination revealed normal ureteral tissue in 10% of the specimens without a crossing vessel and in 43% of the specimens with a crossing vessel. Smooth muscle hypertrophy was found in 27% of patients without a crossing UROLOGY 73 (4), 2009
vessel and in 9% of patients with a crossing vessel. Smooth muscle atrophy was noted in 10% of the specimens without a crossing vessel and in 3% of the specimens with a crossing vessel. Chronic inflammation was seen in 40% of the specimens without a crossing vessel and in 32% of the specimens with a crossing vessel. Fibrosis occurred in 20% of the specimens without a crossing vessel and in 31% of the specimens with a crossing vessel. Overall, patients with a crossing vessel had a statistically significant decrease in observed histologic changes (P ⬍ .0003). As such, patients with a crossing vessel were 7.7-fold less likely to have intrinsic ureteral pathologic features (95% confidence interval 2.1-27.8).
COMMENT The etiology of UPJ obstruction remains unclear in many cases. Multiple mechanisms have been proposed at the structural level, including altered smooth muscle content and/or orientation, abnormal collagen deposition, and abnormalities in UPJ innervation and cellular apoptosis.10 Whether these abnormalities are primary (congenital defects) or secondary (histologic sequela of another primary etiology) is not clear. The role of crossing renal vessels in the pathogenesis of UPJ obstruction is continually debated, and whether crossing vessels are related to histologic/ultrastructural UPJ abnormalities is poorly understood. Several previous studies have investigated the pathologic changes seen at the obstructed UPJ. Normally, the ureter is characterized by a spiral pattern of smooth musculature, casually arranged as a mixture of circular, longitudinal, and oblique bundles, with an inner urothelium consisting of lamina propria and transitional epithelium.8 No microscopic features can distinguish the normal UPJ, with the exception of a preponderance of circular muscle at the UPJ compared with a preponderance of longitudinal muscle more distally.11,12 Histopathologic changes that have been associated with UPJ obstruction include a replacement of normal ureteral musculature with abnormal longitudinal muscle bundles or fibrous tissue and smooth muscle hypotrophy; also, electron microscopy studies have demonstrated excessive collagen deposition.8,9,13,14 Few data exist regarding the relationship of crossing vessels to the prevalence and nature of histopathologic 717
changes in patients with UPJ obstruction. Foote et al.11 reviewed the pathologic findings of 58 patients presenting with UPJ obstruction. A subset of 12 patients was found to have crossing vessels. On pathologic examination, changes were noted only in the musculature of the ureter—including absence of muscle at the UPJ, atrophy of the muscle layer, and hypertrophy of muscle and normal muscle. Of the 12 patients with crossing vessels, one third had muscle atrophy and the other two thirds had muscle hypertrophy. The most common histologic change found in the crossing vessel group was periureteral fibrosis. During surgery, the periureteral and peripelvic fibrosis is dissected to mobilize the crossing vessel from the UPJ. Significant fibrosis surrounding the UPJ could cause mechanical obstruction and, in theory, be independent of the degree of fibrosis seen histologically. However, dense fibrotic tissue surrounding the UPJ was an uncommon finding in our experience. Moreover, we did not appreciate a difference in the degree of gross fibrosis surrounding the UPJ in patients with and without crossing vessels. We have presented a much larger series, including 65 patients with UPJ obstruction and crossing vessels, and argue that crossing vessels do play a role in UPJ obstruction. That 43% of the obstructed patients with associated crossing vessels had no demonstrable histopathologic changes implicates the vessel as the sole, mechanical, etiology of obstruction. In the future, the confirmation of normal UPJ architecture at the ultrastructural (ie, electron microscopic) level would support this hypothesis. The normal histologic findings we observed in this subset of patients does not rule out additional concomitant etiologies of obstruction (eg, neural, molecular). Our histopathologic analysis used a hematoxylin-eosin staining method, which is not optimal to assess for neural or molecular pathologic features. Future prospective, intraoperative functional studies might help elucidate these issues further. Our findings may shed light on the pathogenesis of obstruction caused by crossing vessels. It has been suggested that the difference between obstructing and nonobstructing crossing vessels is the presence and absence, respectively, of UPJ inflammation. Thus, if a crossing vessel leads to UPJ fibrosis and smooth muscle hypertrophy, obstruction ensues. In contrast, our finding that a crossing vessel can be associated with UPJ obstruction without demonstrable histologic changes suggests that such inflammation is not a prerequisite to obstruction and that simple extrinsic compression can cause obstruction in some patients. In other patients, secondary tissue changes might become a factor. This could help explain why the division of crossing vessels does not necessarily relieve obstruction, a finding that has been used to argue against vessels as the source of obstruction.11 In those cases in which significant secondary histologic changes have developed in the UPJ itself, simple vessel ligation is unlikely to be therapeutic. Although the most common histologic observation in the group with crossing vessels was a lack of change in 718
the UPJ tissue, only 10% of patients without crossing vessels had normal UPJ histologic findings. Chronic inflammation was the predominant histologic feature in these patients. These findings support the concept that UPJ obstruction is a heterogeneous disease without a single underlying etiology in all cases. Do these findings have clinical significance? It seems logical that endopyelotomy is more likely to resolve obstruction that is characterized by intrinsic UPJ pathologic features (eg, fibrosis, scarring), but that incision is less likely to affect obstruction caused by external compression alone, without associated intrinsic changes. If true, our finding that 43% of patients with UPJ obstruction and an associated crossing vessel have no intrinsic histologic pathologic findings might help to explain the lower success rates seen with endoincision in patients with crossing vessels.6,7,15,16 However, this hypothesis is challenged by others who have failed to demonstrate a negative effect of crossing vessels on endoincision outcomes.17,18 When a crossing vessel is identified intraoperatively during open or laparoscopic pyeloplasty, its role in the pathophysiology of the functional obstruction can still be called into question. In certain cases, an obvious kinking of the ureter is easily appreciated at the level of the crossing vessel. In others, this is not so easily apparent. In our practice, we have treated select patients in whom a crossing vessel has been identified, but the anatomic relationship of the vessel to the obstructed segment suggested that the vessel was an “innocent bystander.” In such patients, nondismembered pyeloplasty can be considered. However, in our practice, most patients with obstruction and demonstrable crossing vessels undergo dismembered repair.
CONCLUSIONS The results of our study have shown that patients with UPJ obstruction and an associated crossing vessel are less likely to have histopathologic abnormalities compared with patients without crossing vessels. This finding supports the argument that in a subset of patients, crossing vessels cause obstruction by simple extrinsic compression alone, without associated UPJ inflammation.
References 1. Sampaio FJ. The dilemma of the crossing vessel at the ureteropelvic junction: precise anatomic study. J Endourol. 1996;10:411-415. 2. Sampaio FJ. Vascular anatomy at the ureteropelvic junction. Urol Clin North Am. 1998;25:251-258. 3. Zeltser IS, Liu JB, Bagley DH. The incidence of crossing vessels in patients with normal ureteropelvic junction examined with endoluminal ultrasound. J Urol. 2004;172:2304-2307. 4. Sampaio FJ, Favorito LA. Ureteropelvic junction stenosis: vascular anatomical background for endopyelotomy. J Urol. 1993;150:17871791. 5. Bagley DH, Liu JB. Endoureteral sonography to define the anatomy of the obstructed ureteropelvic junction. Urol Clin North Am. 1998;25:271-279. 6. Tawfiek ER, Liu JB, Bagley DH. Ureteroscopic treatment of ureteropelvic junction obstruction. J Urol. 1998;160:1643-1646.
UROLOGY 73 (4), 2009
7. Conlin MJ. Results of selective management of ureteropelvic junction obstruction. J Endourol. 2002;16:233-236. 8. Hanna MK, Jeffs RD, Sturgess JM, et al. Ureteral stricture and ultrastructure. Part I. The normal human ureter. J Urol. 1976;116: 718-724. 9. Hanna MK, Jeffs RD, Sturgess JM, et al. Ureteral structure and ultrastructure. II. Congenital ureteropelvic junction obstruction and primary obstructive megaureter. J Urol. 1976;116:725-730. 10. Kajbafzadeh AM, Payabvash S, Salmasi AH, et al. Smooth muscle cell apoptosis and defective neural development in congenital ureteropelvic junction obstruction. J Urol. 2006;176:718-723. 11. Foote JW, Blennerhassett JB, Wiglesworth FW, et al. Observations on the ureteropelvic junction. J Urol. 1970;104:252-257. 12. Scardino PT, Scardino PL. Obstruction at the ureteropelvic junction. In: Bergman H, ed. The Ureter. New York: Springer-Verlag; 1981:697-717. 13. Ponincasa P, Baroli F, Di Ciaula AD, et al. Defective in-vitro contractility of ureteropelvic junction in children with functional and obstructive urine flow impairment. J Pediatr Surg. 2006;41: 1594-1597. 14. Hsu THS, Streem SB, Nakada SY. Management of upper urinary tract obstruction. In: Wein AJ, Kavoussi LR, Peters CA, et al. eds. Campbell-Walsh Urology, vol. 2. Philadelphia: Saunders Elsevier; 2007:1227-1273. 15. Van Cangh PJ, Wilmart JF, Opsomer RJ, et al. Long-term results and late recurrence after endoureteropyelotomy: a critical analysis of prognostic factors. J Urol. 1994;151:934-937. 16. Conlin MJ, Bagley DH. Ureteroscopic endopyelotomy at a single setting. J Urol. 1998;159:727-731. 17. Gupta M, Tuncay OL, Smith AD. Open surgical exploration after failed endopyelotomy: a 12-year perspective. J Urol. 1997;157: 1613-1618. 18. Gupta M, Smith AD. Crossing vessels at the ureteropelvic junction: do they influence endopyelotomy outcome? J Endourol. 1996;10:183-187.
EDITORIAL COMMENT Advances in imaging techniques have increased the accuracy of diagnosing crossing vessels in patients with ureteropelvic (UPJ) obstruction. Multiple imaging modalities (ie, contrast-enhanced color Doppler ultrasonography, endoluminal ultrasonography, computed tomography, and magnetic resonance angiography) have been found to be very sensitive and specific in predicting the presence of crossing vessels.1,2 From previous studies, we know that the correlation between UPJ obstruction and crossing vessels is real, because patients with UPJ obstruction have a greater incidence of crossing vessels than those with a normal UPJ.1,2 However, the role of these lower pole crossing vessels in the pathophysiology of UPJ obstruction has long been debated. It is still unclear whether the obstruction is from mechanical compression from the crossing vessels or from pathologic/functional changes in the UPJ. This study reports on some interesting findings regarding the pathologic anatomy of the obstructed UPJ. The authors report their retrospective analysis of 155 consecutive patients who underwent laparoscopic pyeloplasty, in which they performed a histopathologic examination of UPJ tissue excised at pyeloplasty. Patients with a crossing vessel had a greater overall incidence of normal pathologic findings (43% vs 10%). It is quite fair to conclude that patients with UPJ obstruction and associated crossing vessels are less likely to have “nonspecific histopathologic abnormalities”; however, it does not prove a causal relationship between the presence of crossing vessels and UPJ obstruction. A prospective intraoperative functional study
UROLOGY 73 (4), 2009
comparing crossing vessels compressing UPJ with crossing vessels retracted off the UPJ might best answer that question. In addition, other concomitant etiologies (eg, neural, molecular mechanisms) might be present that are not assessed with hematoxylin-eosin staining and might play a significant role in causing functional obstruction without any histopathologic abnormalities. Preoperative knowledge of crossing vessels has 2 clinical implications for the urologist. The presence of crossing vessels is associated with a lower success rate after endoscopic incision, regardless of the degree of hydronephrosis and/or renal function.3 Endoscopic treatment in these cases should also be avoided to prevent severe hemorrhage from injury to these crossing vessels. Second, knowledge of the crossing vessels increases intraoperative alertness while dissecting the UPJ during laparoscopic/open pyeloplasty. However, the study concentrates more on the histopathologic aspects of the UPJ and has no clinical relevance in decision making and selecting the appropriate treatment (intraluminal vs extraluminal) for an individual patient with preoperatively identified anatomy (crossing vessel vs no crossing vessel). Gaurav Bandi, M.D., Department of Urology, Jefferson Medical College, Thomas Jefferson University, Philadelphia, Pennsylvania
References 1. Mitterberger M, Pinggera GM, Neururer R, et al. Comparison of contrast-enhanced color Doppler imaging (CDi), computed tomography (CT), and magnetic resonance imaging (MRI) for the detection of crossing vessels in patients with ureteropelvic junction obstruction (UPJO). Eur Urol. 2008;53:1254-1260. 2. Zeltser IS, Liu JB, Bagley DH. The incidence of crossing vessels in patients with normal ureteropelvic junction examined with endoluminal ultrasound. J Urol. 2004;172:2304-2307. 3. Van Cangh PJ, Wilmart JF, Opsomer RJ, et al. Long-term results and late recurrence after endoureteropyelotomy: a critical analysis of prognostic factors. J Urol. 1994;151:934-937.
doi:10.1016/j.urology.2008.11.045 UROLOGY 73: 719, 2009. © 2009 Elsevier Inc.
REPLY Although we thank the authors for the thoughtful reply, we respectfully disagree with the statement that it is “unfortunate” that our report concentrates on the histopathologic findings. In fact, that is the central purpose of our report: to further elucidate the histologic findings encountered intraoperatively in patients undergoing repair for UPJ obstruction and to shed light on the pathophysiology of obstruction. Lee Richstone, M.D., Casey A. Seideman, M.D., Ernesto Reggio, M.D., Rachel Bluebond-Langner, M.D., Peter A. Pinto, M.D., Bruce Trock, M.D., and Louis R. Kavoussi, M.D., Smith Institute for Urology, North Shore-Long Island Jewish Health System, New Hyde Park, New York doi:10.1016/j.urology.2008.12.008 UROLOGY 73: 719, 2009. © 2009 Elsevier Inc.
719
METHODS
RESULTS
CONCLUSIONS
To define the role of crossing vessels in the pathophysiology of ureteropelvic junction (UPJ) obstruction, we analyzed the relationship between the presence of crossing vessels and UPJ pathologic findings in patients undergoing laparoscopic pyeloplasty. The significance of crossing renal vessels in patients with UPJ obstruction is unclear. We performed a retrospective analysis of 155 consecutive patients undergoing laparoscopic pyeloplasty. Pathologic specimens from the UPJ were evaluated in 95 patients. The presence or absence of crossing vessels was documented intraoperatively. The histopathologic findings allowed for categorization into 5 groups: group 1, normal ureteral tissue; group 2, chronic inflammation; group 3, smooth muscle hypertrophy, group 4, fibrosis; and group 5, smooth muscle atrophy. The pathologic findings between patients with and without crossing vessels were compared. Overall, crossing vessels were identified in 98 patients (63.2%). Of the 95 cases with specimens retrieved for histologic analysis, 65 had crossing vessels and 30 did not. The most common UPJ histologic finding in patients with crossing vessels was no intrinsic abnormality (43%). In contrast, this was seen in only 10% of patients without a crossing vessel. In the group without crossing vessels, chronic inflammation (40%) was the predominant histologic findings. Patients with a crossing vessel were less likely to have intrinsic histologic pathologic findings (P ⬍ .0003). Patients with crossing vessels and UPJ obstruction had no histologic abnormalities identified in 43% of cases. This finding implicates crossing vessels in the pathogenesis of select cases of UPJ obstruction and direct mechanical compression as the etiology of obstruction in these individuals. UROLOGY 73: 716 –719, 2009. © 2009 Published by Elsevier Inc.
B
oth intrinsic and extrinsic factors have been associated with ureteropelvic junction (UPJ) obstruction, including intrinsic structural abnormalities, lower pole crossing renal vessels, external mass compression, retroperitoneal inflammation, abdominal aortic aneurysm, calculi, and infection. Obstruction at the UPJ, regardless of the etiology, can result in hydronephrosis, pain, and loss of renal function. The role of crossing renal vessels in patients with UPJ obstruction has been the subject of debate.1-3 The true incidence of lower pole vessels in the general (unobstructed) population is not known. In cadaveric studies, Sampaio1,2 and Sampaio and Favorito4 demonstrated crossing vessels in ⱕ71% of patients dying of various causes; however, it is unclear how accurately this reflects in vivo anatomic relationships. Imaging studies have From the Smith Institute for Urology, North Shore-Long Island Jewish Health System, New Hyde Park, New York Reprint requests: Lee Richstone, M.D., Smith Institute for Urology, North ShoreLong Island Jewish Health System, 450 Lakeville Road, Suite M41, New Hyde Park, NY 11040. E-mail: [email protected] Submitted: April 3, 2008, accepted (with revisions): October 9, 2008
716
© 2009 Published by Elsevier Inc.
suggested an association between crossing vessels and UPJ obstruction. Endoluminal ultrasound studies have demonstrated crossing vessels at the UPJ in 19.2% of patients with a normal UPJ but in 53%-71% of patients with UPJ obstruction.3,5-7 This has been confirmed by additional radiographic studies using conventional angiography, spiral computed tomography angiography, and contrast-enhanced Doppler ultrasonography, documenting crossing vessels in 39%-79% of patients with UPJ obstruction.3 Despite this association, the definitive causal relationship, if any, between crossing vessels and UPJ obstruction remains poorly understood. Numerous studies have attempted to unravel the pathophysiology of UPJ obstruction by histologic evaluation of UPJ tissue.8,9 Although data exist describing the histologic features of the UPJ in patients with UPJ obstruction using both light and electron microscopy,8,9 few if any reports have specifically addressed the histologic findings in patients with and without crossing vessels. Such information has the potential to shed light on the role of crossing vessels in the pathogenesis of UPJ obstruction. We analyzed the histopathologic UPJ specimens of a large series 0090-4295/09/$34.00 doi:10.1016/j.urology.2008.10.069
Table 1. Results of pathologic examination Pathologic Finding
No Crossing Vessel (n ⫽ 30)
Crossing Vessel (n ⫽ 65)
10 (3) 90 (27) 27 (7/27) 40 (11/27) 20 (6/27) 20 (6/27)
43 (28) 57 (37) 9 (3/37) 32 (13/37) 31 (12/37) 31 (12/37)
Normal Abnormal with significant overlap of following histopathologic findings SM hypertrophy Chronic inflammation Fibrosis SM atrophy SM ⫽ smooth muscle. Data presented as percentages, with numbers in parentheses.
of patients with primary UPJ obstruction in an attempt to better understand the mechanisms by which crossing renal vessels might be involved in obstruction.
MATERIAL AND METHODS We performed a retrospective review of 155 consecutive patients who underwent laparoscopic pyeloplasty for primary UPJ obstruction by a single surgeon (L.R.K.). All patients had radiographic evidence of obstruction with symptoms and/or deterioration of renal function. All patients were preoperatively evaluated using computed tomography and renal scanning before surgical intervention. All patients demonstrated moderate to severe hydronephrosis at the evaluation. Additionally, all patients had a renal function of ⱖ20%, with a half-time of ⱖ20 minutes. The mean age was 38 years (range 10-85). No statistically significant correlation was found between the pathologic findings and age. Of the 155 patients, 84 were female and 71 were male. The site of UPJ obstruction was the left in 61 patients (39.4%) and the right in 94 (60.6%). The presence or absence of crossing vessels was documented intraoperatively. Crossing vessels were identified in 98 patients (63.2%). Of those with crossing vessels, 13 underwent nondismembered repair and were excluded from the analysis; all patients for whom pathologic evaluation was done had undergone Anderson-Hynes dismembered pyeloplasty. All patients had stents placed intraoperatively. The patients in whom previous pyeloplasty, antegrade or retrograde endopyelotomy had failed and those with evidence of secondary obstruction were excluded at the discretion of the operating surgeon. This left 95 patients available for analysis in whom the pathologic findings were evaluated, including 65 with crossing vessels and 30 without. Microscopic evaluation was performed with standard hematoxylin-eosin staining techniques. Fischer’s exact t test was used to calculate the P values.
RESULTS Five histologic categories were identified: (a) normal ureteral tissue, (b) chronic inflammation, (c) smooth muscle hypertrophy, (d) fibrosis, and (e) smooth muscle atrophy. The specimens that had abnormal histologic characteristics demonstrated significant overlap in the findings. The results are listed in Table 1. Histologic examination revealed normal ureteral tissue in 10% of the specimens without a crossing vessel and in 43% of the specimens with a crossing vessel. Smooth muscle hypertrophy was found in 27% of patients without a crossing UROLOGY 73 (4), 2009
vessel and in 9% of patients with a crossing vessel. Smooth muscle atrophy was noted in 10% of the specimens without a crossing vessel and in 3% of the specimens with a crossing vessel. Chronic inflammation was seen in 40% of the specimens without a crossing vessel and in 32% of the specimens with a crossing vessel. Fibrosis occurred in 20% of the specimens without a crossing vessel and in 31% of the specimens with a crossing vessel. Overall, patients with a crossing vessel had a statistically significant decrease in observed histologic changes (P ⬍ .0003). As such, patients with a crossing vessel were 7.7-fold less likely to have intrinsic ureteral pathologic features (95% confidence interval 2.1-27.8).
COMMENT The etiology of UPJ obstruction remains unclear in many cases. Multiple mechanisms have been proposed at the structural level, including altered smooth muscle content and/or orientation, abnormal collagen deposition, and abnormalities in UPJ innervation and cellular apoptosis.10 Whether these abnormalities are primary (congenital defects) or secondary (histologic sequela of another primary etiology) is not clear. The role of crossing renal vessels in the pathogenesis of UPJ obstruction is continually debated, and whether crossing vessels are related to histologic/ultrastructural UPJ abnormalities is poorly understood. Several previous studies have investigated the pathologic changes seen at the obstructed UPJ. Normally, the ureter is characterized by a spiral pattern of smooth musculature, casually arranged as a mixture of circular, longitudinal, and oblique bundles, with an inner urothelium consisting of lamina propria and transitional epithelium.8 No microscopic features can distinguish the normal UPJ, with the exception of a preponderance of circular muscle at the UPJ compared with a preponderance of longitudinal muscle more distally.11,12 Histopathologic changes that have been associated with UPJ obstruction include a replacement of normal ureteral musculature with abnormal longitudinal muscle bundles or fibrous tissue and smooth muscle hypotrophy; also, electron microscopy studies have demonstrated excessive collagen deposition.8,9,13,14 Few data exist regarding the relationship of crossing vessels to the prevalence and nature of histopathologic 717
changes in patients with UPJ obstruction. Foote et al.11 reviewed the pathologic findings of 58 patients presenting with UPJ obstruction. A subset of 12 patients was found to have crossing vessels. On pathologic examination, changes were noted only in the musculature of the ureter—including absence of muscle at the UPJ, atrophy of the muscle layer, and hypertrophy of muscle and normal muscle. Of the 12 patients with crossing vessels, one third had muscle atrophy and the other two thirds had muscle hypertrophy. The most common histologic change found in the crossing vessel group was periureteral fibrosis. During surgery, the periureteral and peripelvic fibrosis is dissected to mobilize the crossing vessel from the UPJ. Significant fibrosis surrounding the UPJ could cause mechanical obstruction and, in theory, be independent of the degree of fibrosis seen histologically. However, dense fibrotic tissue surrounding the UPJ was an uncommon finding in our experience. Moreover, we did not appreciate a difference in the degree of gross fibrosis surrounding the UPJ in patients with and without crossing vessels. We have presented a much larger series, including 65 patients with UPJ obstruction and crossing vessels, and argue that crossing vessels do play a role in UPJ obstruction. That 43% of the obstructed patients with associated crossing vessels had no demonstrable histopathologic changes implicates the vessel as the sole, mechanical, etiology of obstruction. In the future, the confirmation of normal UPJ architecture at the ultrastructural (ie, electron microscopic) level would support this hypothesis. The normal histologic findings we observed in this subset of patients does not rule out additional concomitant etiologies of obstruction (eg, neural, molecular). Our histopathologic analysis used a hematoxylin-eosin staining method, which is not optimal to assess for neural or molecular pathologic features. Future prospective, intraoperative functional studies might help elucidate these issues further. Our findings may shed light on the pathogenesis of obstruction caused by crossing vessels. It has been suggested that the difference between obstructing and nonobstructing crossing vessels is the presence and absence, respectively, of UPJ inflammation. Thus, if a crossing vessel leads to UPJ fibrosis and smooth muscle hypertrophy, obstruction ensues. In contrast, our finding that a crossing vessel can be associated with UPJ obstruction without demonstrable histologic changes suggests that such inflammation is not a prerequisite to obstruction and that simple extrinsic compression can cause obstruction in some patients. In other patients, secondary tissue changes might become a factor. This could help explain why the division of crossing vessels does not necessarily relieve obstruction, a finding that has been used to argue against vessels as the source of obstruction.11 In those cases in which significant secondary histologic changes have developed in the UPJ itself, simple vessel ligation is unlikely to be therapeutic. Although the most common histologic observation in the group with crossing vessels was a lack of change in 718
the UPJ tissue, only 10% of patients without crossing vessels had normal UPJ histologic findings. Chronic inflammation was the predominant histologic feature in these patients. These findings support the concept that UPJ obstruction is a heterogeneous disease without a single underlying etiology in all cases. Do these findings have clinical significance? It seems logical that endopyelotomy is more likely to resolve obstruction that is characterized by intrinsic UPJ pathologic features (eg, fibrosis, scarring), but that incision is less likely to affect obstruction caused by external compression alone, without associated intrinsic changes. If true, our finding that 43% of patients with UPJ obstruction and an associated crossing vessel have no intrinsic histologic pathologic findings might help to explain the lower success rates seen with endoincision in patients with crossing vessels.6,7,15,16 However, this hypothesis is challenged by others who have failed to demonstrate a negative effect of crossing vessels on endoincision outcomes.17,18 When a crossing vessel is identified intraoperatively during open or laparoscopic pyeloplasty, its role in the pathophysiology of the functional obstruction can still be called into question. In certain cases, an obvious kinking of the ureter is easily appreciated at the level of the crossing vessel. In others, this is not so easily apparent. In our practice, we have treated select patients in whom a crossing vessel has been identified, but the anatomic relationship of the vessel to the obstructed segment suggested that the vessel was an “innocent bystander.” In such patients, nondismembered pyeloplasty can be considered. However, in our practice, most patients with obstruction and demonstrable crossing vessels undergo dismembered repair.
CONCLUSIONS The results of our study have shown that patients with UPJ obstruction and an associated crossing vessel are less likely to have histopathologic abnormalities compared with patients without crossing vessels. This finding supports the argument that in a subset of patients, crossing vessels cause obstruction by simple extrinsic compression alone, without associated UPJ inflammation.
References 1. Sampaio FJ. The dilemma of the crossing vessel at the ureteropelvic junction: precise anatomic study. J Endourol. 1996;10:411-415. 2. Sampaio FJ. Vascular anatomy at the ureteropelvic junction. Urol Clin North Am. 1998;25:251-258. 3. Zeltser IS, Liu JB, Bagley DH. The incidence of crossing vessels in patients with normal ureteropelvic junction examined with endoluminal ultrasound. J Urol. 2004;172:2304-2307. 4. Sampaio FJ, Favorito LA. Ureteropelvic junction stenosis: vascular anatomical background for endopyelotomy. J Urol. 1993;150:17871791. 5. Bagley DH, Liu JB. Endoureteral sonography to define the anatomy of the obstructed ureteropelvic junction. Urol Clin North Am. 1998;25:271-279. 6. Tawfiek ER, Liu JB, Bagley DH. Ureteroscopic treatment of ureteropelvic junction obstruction. J Urol. 1998;160:1643-1646.
UROLOGY 73 (4), 2009
7. Conlin MJ. Results of selective management of ureteropelvic junction obstruction. J Endourol. 2002;16:233-236. 8. Hanna MK, Jeffs RD, Sturgess JM, et al. Ureteral stricture and ultrastructure. Part I. The normal human ureter. J Urol. 1976;116: 718-724. 9. Hanna MK, Jeffs RD, Sturgess JM, et al. Ureteral structure and ultrastructure. II. Congenital ureteropelvic junction obstruction and primary obstructive megaureter. J Urol. 1976;116:725-730. 10. Kajbafzadeh AM, Payabvash S, Salmasi AH, et al. Smooth muscle cell apoptosis and defective neural development in congenital ureteropelvic junction obstruction. J Urol. 2006;176:718-723. 11. Foote JW, Blennerhassett JB, Wiglesworth FW, et al. Observations on the ureteropelvic junction. J Urol. 1970;104:252-257. 12. Scardino PT, Scardino PL. Obstruction at the ureteropelvic junction. In: Bergman H, ed. The Ureter. New York: Springer-Verlag; 1981:697-717. 13. Ponincasa P, Baroli F, Di Ciaula AD, et al. Defective in-vitro contractility of ureteropelvic junction in children with functional and obstructive urine flow impairment. J Pediatr Surg. 2006;41: 1594-1597. 14. Hsu THS, Streem SB, Nakada SY. Management of upper urinary tract obstruction. In: Wein AJ, Kavoussi LR, Peters CA, et al. eds. Campbell-Walsh Urology, vol. 2. Philadelphia: Saunders Elsevier; 2007:1227-1273. 15. Van Cangh PJ, Wilmart JF, Opsomer RJ, et al. Long-term results and late recurrence after endoureteropyelotomy: a critical analysis of prognostic factors. J Urol. 1994;151:934-937. 16. Conlin MJ, Bagley DH. Ureteroscopic endopyelotomy at a single setting. J Urol. 1998;159:727-731. 17. Gupta M, Tuncay OL, Smith AD. Open surgical exploration after failed endopyelotomy: a 12-year perspective. J Urol. 1997;157: 1613-1618. 18. Gupta M, Smith AD. Crossing vessels at the ureteropelvic junction: do they influence endopyelotomy outcome? J Endourol. 1996;10:183-187.
EDITORIAL COMMENT Advances in imaging techniques have increased the accuracy of diagnosing crossing vessels in patients with ureteropelvic (UPJ) obstruction. Multiple imaging modalities (ie, contrast-enhanced color Doppler ultrasonography, endoluminal ultrasonography, computed tomography, and magnetic resonance angiography) have been found to be very sensitive and specific in predicting the presence of crossing vessels.1,2 From previous studies, we know that the correlation between UPJ obstruction and crossing vessels is real, because patients with UPJ obstruction have a greater incidence of crossing vessels than those with a normal UPJ.1,2 However, the role of these lower pole crossing vessels in the pathophysiology of UPJ obstruction has long been debated. It is still unclear whether the obstruction is from mechanical compression from the crossing vessels or from pathologic/functional changes in the UPJ. This study reports on some interesting findings regarding the pathologic anatomy of the obstructed UPJ. The authors report their retrospective analysis of 155 consecutive patients who underwent laparoscopic pyeloplasty, in which they performed a histopathologic examination of UPJ tissue excised at pyeloplasty. Patients with a crossing vessel had a greater overall incidence of normal pathologic findings (43% vs 10%). It is quite fair to conclude that patients with UPJ obstruction and associated crossing vessels are less likely to have “nonspecific histopathologic abnormalities”; however, it does not prove a causal relationship between the presence of crossing vessels and UPJ obstruction. A prospective intraoperative functional study
UROLOGY 73 (4), 2009
comparing crossing vessels compressing UPJ with crossing vessels retracted off the UPJ might best answer that question. In addition, other concomitant etiologies (eg, neural, molecular mechanisms) might be present that are not assessed with hematoxylin-eosin staining and might play a significant role in causing functional obstruction without any histopathologic abnormalities. Preoperative knowledge of crossing vessels has 2 clinical implications for the urologist. The presence of crossing vessels is associated with a lower success rate after endoscopic incision, regardless of the degree of hydronephrosis and/or renal function.3 Endoscopic treatment in these cases should also be avoided to prevent severe hemorrhage from injury to these crossing vessels. Second, knowledge of the crossing vessels increases intraoperative alertness while dissecting the UPJ during laparoscopic/open pyeloplasty. However, the study concentrates more on the histopathologic aspects of the UPJ and has no clinical relevance in decision making and selecting the appropriate treatment (intraluminal vs extraluminal) for an individual patient with preoperatively identified anatomy (crossing vessel vs no crossing vessel). Gaurav Bandi, M.D., Department of Urology, Jefferson Medical College, Thomas Jefferson University, Philadelphia, Pennsylvania
References 1. Mitterberger M, Pinggera GM, Neururer R, et al. Comparison of contrast-enhanced color Doppler imaging (CDi), computed tomography (CT), and magnetic resonance imaging (MRI) for the detection of crossing vessels in patients with ureteropelvic junction obstruction (UPJO). Eur Urol. 2008;53:1254-1260. 2. Zeltser IS, Liu JB, Bagley DH. The incidence of crossing vessels in patients with normal ureteropelvic junction examined with endoluminal ultrasound. J Urol. 2004;172:2304-2307. 3. Van Cangh PJ, Wilmart JF, Opsomer RJ, et al. Long-term results and late recurrence after endoureteropyelotomy: a critical analysis of prognostic factors. J Urol. 1994;151:934-937.
doi:10.1016/j.urology.2008.11.045 UROLOGY 73: 719, 2009. © 2009 Elsevier Inc.
REPLY Although we thank the authors for the thoughtful reply, we respectfully disagree with the statement that it is “unfortunate” that our report concentrates on the histopathologic findings. In fact, that is the central purpose of our report: to further elucidate the histologic findings encountered intraoperatively in patients undergoing repair for UPJ obstruction and to shed light on the pathophysiology of obstruction. Lee Richstone, M.D., Casey A. Seideman, M.D., Ernesto Reggio, M.D., Rachel Bluebond-Langner, M.D., Peter A. Pinto, M.D., Bruce Trock, M.D., and Louis R. Kavoussi, M.D., Smith Institute for Urology, North Shore-Long Island Jewish Health System, New Hyde Park, New York doi:10.1016/j.urology.2008.12.008 UROLOGY 73: 719, 2009. © 2009 Elsevier Inc.
719