- Email: [email protected]
EARLY DIURESIS RENOGRAM FINDINGS PREDICT SUCCESS FOLLOWING PYELOPLASTY
0022-5347/01/1656-2311/0 THE JOURNAL OF UROLOGY® Copyright © 2001 by AMERICAN UROLOGICAL ASSOCIATION, INC.®
Vol. 165, 2311–2315, June 2001 Printed in U.S.A.
EARLY DIURESIS RENOGRAM FINDINGS PREDICT SUCCESS FOLLOWING PYELOPLASTY HANS G. POHL, H. GIL RUSHTON, JAE-SHIN PARK, A. BARRY BELMAN
AND
MASSOUD MAJD
From the Departments of Pediatric Urology and Nuclear Medicine, Children’s National Medical Center, Washington, D. C.
ABSTRACT
Purpose: We sought to determine whether drainage across the ureteropelvic junction, as indicated by diuretic renography 3 months after pyeloplasty, is an adequate predictor of surgical success. Materials and Methods: The medical records of 150 children who underwent pyeloplasty from 1986 to 1995 were reviewed. After excluding nonevaluable cases a total of 127 renal units remained for investigation. Preoperatively each renal unit was examined with a standardized (well-tempered) furosemide stimulated renal scan. Postoperatively 60 renal units were evaluated with standardized diuretic renal scans at 3 and 12 months, 33 renal units at 3 months only and 34 renal units at 12 months only. Surgical success was defined by half-time less than 20 minutes on a standardized diuretic renogram. Results: Of the 33 renal units with a single postoperative study at 3 months 32 (97%) had halftime less than 20 minutes on diuretic renography. The remaining patient in this group with half-time greater than 20 minutes showed 60% improvement in half-time and did not require reoperation. Excluding those without delayed followup, surgical success was obtained in 93 of the 94 (99%) renal units. Among the 60 renal units evaluated with 2 postoperative renal scans success was noted in 48 (80%) and 59 (98%) at 3 and 12 months, respectively. Stenosis did not recur in 48 renal units with half-time less than 20 minutes 3 months after repair. In 1 case that had been treated for postoperative urinoma half-time was greater than 40 minutes at 3 months and repeat pyeloplasty was required. Conclusions: Half-time less than 20 minutes 3 months after pyeloplasty predicts surgical success. Most renal units that improve but still have half-times greater than 20 minutes on an early diuretic renogram will demonstrate continued improvement in drainage patterns at 12 months. Those renal units that show no improvement at 3 months may require reoperation and those with half-time less than 20 minutes at 3 months do not require further evaluation. KEY WORDS: diuresis, radioisotope renography, kidney
Diuretic renography remains the most objective means of assessing renal function and drainage associated with ureteropelvic junction obstruction,1 and a valid means of predicting postoperative success. It has been our impression that urinary drainage after successful repair of ureteropelvic junction obstruction has improved significantly by 3 months with little further improvement at 12 months. Therefore, we hypothesized that normalization of diuretic renography at 3 months postoperatively might obviate the need for long-term imaging at 1 year. Additionally, we propose that diuretic renography offers optimal assessment after pyeloplasty as a consequence of its ability to measure quantitatively relative renal function and urinary transit across the ureteropelvic junction obstruction. As a means of comparison we review reports of the use of excretory urography (IVP) and renal ultrasonography for post-pyeloplasty evaluation.
TABLE 1. Demographic data on 150 patients treated surgically for ureteropelvic junction obstruction % (No.) Gender: Males Females Race:* White Black Hispanic Asian Laterality:† Lt. Rt. Bilat. Pathology: Stenotic ureteropelvic junction obstruction Crossing vessel Fibroepithelial polyp * Race data not recorded in 11 patients. † Laterality not recorded in 8 patients.
MATERIALS AND METHODS
The medical records of 150 children (160 hydronephrotic renal units) who underwent pyeloplasty for ureteropelvic junction obstruction between 1986 and 1995 were reviewed. A total of 33 patients (33 units) were excluded from study because of insufficient renographic data (table 1). The remaining 117 patients (127 renal units) had undergone preoperative and postoperative standardized furosemide stimulated renal scans and divided into 1 of 3 groups according to the timing of postoperative followup (table 2). In group 1 postoperative scans were obtained at 3 and 12
65 (98) 35 (52) 71 (98) 19 (26) 7 (10) 3 (5) 55 (78) 38 (52) 7 (10) 87 (111) 12 (15) 1 (1)
months following repair in 60 cases. To determine whether the preoperative degree of obstruction influenced the rate of postoperative improvement in drainage half-times, the renal units in group 1 were further stratified using preoperative half-times of mild—20 to 40 minutes (23) moderate— 40 to 70 minutes (10) and severe— greater than 70 minutes (27). Group 2 consisted of 33 kidneys with complete furosemide renographic data only at 3 months after pyeloplasty. Of these
2311
2312
EARLY DIURESIS RENOGRAM FINDINGS PREDICT SUCCESS FOLLOWING PYELOPLASTY
TABLE 2. Surgical success following pyeloplasty for 127 renal units divided by the availability of postoperative renographic data Postop.
Group No.
Preop.
Early (3 mos.)
Late (12 mos.)
1 2 3
X X X
X X
X
% Success (No./total No.) 98 (59/60) 97 (32/33) 100 (34/34)
X
units 19 were lost to followup, repeat renography was done in 9 at 12 months but no half-time data were recorded and 5 units were evaluated with ultrasonography at 12 months rather than nuclear scintigraphy. Group 3 consisted of 34 cases with complete furosemide renographic data only at 12 months. For the purpose of this study, surgical success was defined as half-time less than 20 minutes postoperatively on a diuresis renogram. RESULTS
Of the group 1 renal units surgical success was demonstrated in 80% at 3 months and 98% at 12 months (table 3). Figure 1 illustrates the drainage patterns that typify the progressive improvement following successful pyeloplasty. When stratified as mild, moderate or severe ureteropelvic junction obstruction based on preoperative half-time values, improvement in half-time to less than 20 minutes at 3 months was noted in 90%, 80% and 70%, respectively (fig. 2). Those with postoperative half-time greater than 20 minutes demonstrated varying degrees of improvement in drainage compared with the preoperative renal scan, and relative function remained stable or improved (fig. 3). At 1 year improvement to half-time less than 20 minutes occurred in 100%, 100% and 96% (26 of 27), respectively, for cases of mild, moderate and severe obstruction, respectively (fig. 2). In 1 patient in the severe group a urinoma developed postoperatively and half-time was greater than 40 minutes at 3 months. At 12 months following initial repair worsening drainage evidenced by half-time greater than 100 minutes prompted a second pyeloplasty. Within 3 months of the repeat procedure half-time was 6 minutes. No patient with half-time less than 20 minutes at 3 months had late obstruction at 1 year. Of the 33 group 2 renal units 32 (97%) had half-time less than 20 minutes on diuretic renography. The patient with half-time greater than 20 minutes showed 60% improvement halftime and, to our knowledge, had no further difficulty (fig. 4). When groups 1 and 2 were combined 86% (80 of 93 units) of all renal units demonstrated half-time less than 20 minutes at 3 months postoperatively. Group 3 renal units had half-
FIG. 1. 99mTechnetium diethylenetetramine-pentoacetic acid diuresis renography of male infant presenting with antenatal hydronephrosis. A, preoperatively significant ureteropelvic junction obstruction is demonstrated by half-time greater than 100 minutes. B, 3 months after pyeloplasty half-time is significantly improved to less than 20 minutes. C, further improvement in drainage half-time is noted 12 months after surgery.
TABLE 3. Comparative results from previous reports of surgical outcome after Anderson-Hynes dismembered pyeloplasty with an emphasis on the method of postoperative followup % Improved at Interval From Surgery Method Followup
IVP: Williams and Kenawi6 Ultrasonography: Neste et al8 al10
No. Cases
1 Wk.
1 Mo.
3 Mos.
176
24 Mos.
Greater Than 24 Mos.
43† 61‡ 34
58† 74‡ 60 94
68† 84‡ 81
72† 96‡ 91
83
91
66.5*
46
47 8 Amling et 38 Tapia and Gonzalez11 14 37 81 92 Nephrostogram: Miyamoto and Mesrobian Renogram: 8 46 Neste et al 38 Tapia and Gonzalez11 Present study * Only minimal improvement in 46%. † SFU hydronephrosis grade used to assess postoperative improvement. ‡ Degree of renal pelvic dilatation used to assess postoperative improvement. § Includes all 93 renal units were evaluated at 3 months. 㛳 Includes all 94 renal units were evaluated at 12 months.
12 Mos.
6 Mos.
21
73 94 86§
99㛳
EARLY DIURESIS RENOGRAM FINDINGS PREDICT SUCCESS FOLLOWING PYELOPLASTY
FIG. 2. Success following pyeloplasty for 60 renal units evaluated with early and late postoperative diuretic renogram. Preoperative half-times were used to categorize renal units into 1 of 3 severity groups.
FIG. 3. Drainage patterns of 11 renal units evaluated by diuretic renography preoperatively and at 3 and 12 months postoperatively. This subset of group 2 includes all renal units that demonstrated half-times greater than 20 minutes 3 months after pyeloplasty. (Broken lines connect half-times of 3 instructive cases.) Of 11 renal units with early half-times greater than 20 minutes 9 (82%) demonstrated progressive improvement in drainage half-times to less than 20 minutes at 12 months (square). Additional renal unit improved to half-time 4 minutes 12 months after pyeloplasty, although diuretic renography at 3 months demonstrated persistently obstructed pattern (diamond). Remaining renal unit required repeat pyeloplasty to repair re-stenosis following urinoma (circle).
time less than 20 minutes on diuretic renography. Of all renal units from which renographic data were available 1 year postoperative (groups 1 and 3 combined) 99% (93 of 94 units) had half-time less than 20 minutes. When all 3 groups were considered 98.4% (125 of 127 units) had postoperative half-time less than 20 minutes at last followup. DISCUSSION
A number of methods have been applied to measure the postoperative results of pyeloplasty, including clinical status, IVP, ultrasonography and diuretic renography. Despite improvement in symptomatology following pyeloplasty, radiographic improvement in hydronephrosis has often been lacking.2– 4 Postoperative IVPs have routinely demonstrated
2313
FIG. 4. Preoperative and postoperative half-times from 30 group 2 renal units evaluated with diuretic renogram 3 months after repair. Solid line demonstrates 60% improvement in drainage pattern of only renal unit in group 2 with half-time greater than 20 minutes on “early” renogram. All remaining 29 renal units had half-times less than 20 minutes (hatched line). One case with half-time less than 20 minutes and biphasic time-activity curve consistent with significant flow dependent obstruction underwent dismembered pyeloplasty.
persistent caliceal clubbing and renal pelvic dilatation even on studies performed a decade after repair.3, 5 In fact, Roberts et al reported that caliceal clubbing improved in only 15 of 75 children (20%) after surgical correction.5 Similarly, in their review of 190 renal units treated with pyeloplasty Williams and Kenawi noted that calicectasis was unchanged or minimally improved in 79% of the cases.6 Only those renal units with minimal calicectasis preoperatively (10%) had significant improvement in caliceal clubbing on followup IVP. In general, when significant improvement was noted on IVP, it most commonly took the form of a reduction in renal pelvic dilatation while calicectasis persisted, which may have been a consequence of excision of a redundant renal pelvis. Stein et al reported similar findings in 193 renal units treated with dismembered pyeloplasty, of which 40 had grade 4 or “giant” hydronephrosis and a dilated renal pelvis that crossed the midline of the abdomen.7 In none of the renal units was the dilated pelvis excised in conjunction with resection of the stenotic ureteropelvic junction obstruction. Of the 40 renal pelves 24 (60%) improved to grade 1 or 2 on the last followup IVP. Despite symptomatic improvement in the remaining 40%, radiographic improvement was minimal or nonexistent (30% grade 3, 10% grade 4). These data suggest that the degree of pyelocalicectasis as demonstrated on IVP is an inaccurate measure of drainage. Additionally, the IVP affords only qualitative data on differential renal function. In 1993 Neste et al published post-pyeloplasty comparison between ultrasonography and diuretic renography of 56 renal units.8 Each unit was evaluated with ultrasonography and diuretic renography preoperatively and multiple times within the year following repair. Sonographic studies were graded according to the Society for Fetal Urology (SFU) classification criteria.9 Surgical success, defined as resolution of flank pain, absence of urinary infection and interval renal growth, occurred in 46 (82%) of the 56 renal units. The remaining 10 renal units associated with persistent flank pain, urinary infection and poor renal growth were deemed “unfavorable surgical outcomes” and were analyzed separately. Of the group that was clinically improved dilatation diminished in only 43% (12 of 28 units) of those evaluated with ultrasonography at 6 months (table 3). With longer intervals following surgery a greater percentage of renal units demonstrated diminished dilatation sonographically. By 24 months following pyeloplasty 72% (33 of 46 units) of those with clinical improvement demonstrated less dilatation. Among the 10 units with unfavorable outcomes 7 were unchanged and 2 demonstrated diminished hydronephrosis
2314
EARLY DIURESIS RENOGRAM FINDINGS PREDICT SUCCESS FOLLOWING PYELOPLASTY
on followup ultrasonography. Neste et al also investigated renal pelvic size alone, measured by greatest diameter in mm., as a predictor of surgical outcome. They reported initial improvement in 61% (17 of 28 units) at 6 months compared with 43% improvement noted when the SFU classification was used. Again, this difference is most likely secondary to surgical reduction of the renal pelvis. With longer followup to 18 months an additional 35% of the renal pelves demonstrated reduced diameters. Similarly, in a study of 44 children (47 kidneys) only 38% showed improvement on ultrasonography 6 months after pyeloplasty (table 3).10 Eventually, 81% demonstrated improvement according to sonographic criteria at 2 years. Thus, these data corroborate the finding of Stein et al that modest improvement in renal pelvic dilatation can be anticipated with time following successful pyeloplasty. Patient age at repair may have a role in the eventual resolution of pyelocalicectasis as noted on sonography or IVP. In the studies reported by Williams and Kenawi,6 and Stein et al,7 in which minimal improvement in calicectasis was noted, mean patient ages were 6.2 and 4.9 years, respectively. In contrast, in a younger cohort (mean age 2.5 years) Tapia and Gonzalez reported a reduction in the grade of hydronephrosis (by SFU classification) in 36 (94%) of the 38 renal units studied by ultrasonography 1 year after pyeloplasty (table 3).11 Despite significant hydronephrosis in all of the renal units treated (grade 3 in 10, grade 4 in 28), calicectasis eventually resolved completely in 75% of the kidneys. Since most ureteropelvic junction obstruction is now detected prenatally and repair is being performed at an earlier age, one might predict an increased number of renal units demonstrating sonographic improvement following pyeloplasty. However, sonographic data from Stein, Neste and Amling et al do not demonstrate that the infant kidney is more likely to have early resolution of pyelocalicectasis postoperatively.7, 8, 10 Since it is essential to detect persistent obstruction early to prevent irreversible loss of renal function, IVP and ultrasonography do not provide adequate assessment postoperatively. In contrast, diuretic renography offers a truly quantifiable assessment of renal function and drainage following repair that can be objectively compared with the preoperative scan. Consequently, postoperative obstruction can be detected promptly allowing for earlier intervention. The technique of diuretic renography has been standardized to minimize inter-examination variability. The standardized approach includes intravenous hydration with 15 ml./kg. dextrose 5%/0.3% normal saline beginning before tracer injection and continuing throughout the study, and placement of a bladder catheter.12 Furosemide (1 mg./kg., maximum 40 mg.) is administered only after maximal accumulation of the radiopharmaceutical has occurred in the dilated collecting system. Despite the commonly held misconception that the neonatal kidney is too immature to excrete radioisotopes and respond appropriately to furosemide challenge, diuretic renography has been shown to be reliable in the assessment of drainage and function in the infant kidney, even in patients as young as 3 days.1 Although it has been demonstrated that a period of healing is required between surgical repair and mucosal bridging (3 weeks), return of normal electrical conductivity (28 days) and completion of smooth muscle bridging (6 weeks), it has been demonstrated that the ureteropelvic junction obstruction can efficiently transmit a bolus of urine as early as 1 week following repair.13–15 Thus, it is most likely that improvement in drainage half-time is an early indicator of surgical outcome and precedes the morphological changes anticipated following relief of obstruction. In contemporary series success rates following pyeloplasty are reported to be between 72% and 98%.8, 10, 11, 14, 16 However, the results of postoperative evaluation vary with the
method of imaging study used and the timing of evaluation (table 3). Surgical success determined by diuretic renography appears to precede sonographic improvement by 9 months as demonstrated by the 73% to 94% success rate noted on renography at 3 months compared to 21% noted on ultrasonography (table 3). Not until 12 months have elapsed does the success rate as demonstrated on ultrasonography (58% to 94%) begin to approximate that observed on diuretic renography (91% to 98%). Our results concur with previously published findings of significant improvement in drainage by 3 months postoperatively, at which time 86% (80 of 93, groups 1 and 2 combined) of our patients had half-time less than 20 minutes.8, 11 We have also shown that the rate of early improvement following pyeloplasty correlates with the degree of preoperative obstruction based on diuretic renogram halftimes. Even so, the majority of kidneys (70%) with even severe obstruction preoperatively (half-time greater than 70 minutes) will by 3 months postoperatively demonstrate surgical success as defined by half-time less than 20 minutes. We acknowledge the possibility that stenosis could have recurred late in group 2 since delayed evaluation was not performed. However, it would be unlikely that this group alone had a significant proportion of late postoperative complications that was shielded from analysis. Furthermore, of those patients who were evaluated with diuretic renography at 3 and 12 months postoperatively none with half-time less than 20 minutes at 3 months demonstrated subsequent deterioration of drainage or late re-stenosis. In fact, 10 of 11 (91%) renal units with half-time greater than 20 minutes at 3 months postoperatively improved to half-time less than 20 minutes at 12 months. Therefore, the overall success rate for cases with followup at 1 year was 99% (93 of 94). The failure occurred in a patient with a postoperative urinoma who ultimately required repeat pyeloplasty for worsening drainage. One could argue that improvement on postoperative ultrasound at 3 months might obviate the need for a diuretic renogram. However, one would sacrifice the opportunity to evaluate postoperative function. Furthermore, since most sonograms are not performed under standardized hydration, the results could be misleading. CONCLUSIONS
When significant improvement is demonstrated on diuretic renography 3 months after dismembered pyeloplasty (halftime less than 20 minutes) no further followup is required. However, it does seem prudent to follow renal units with half-times greater than 20 minutes and no functional loss with additional diuretic renography performed 9 to 12 months after pyeloplasty. Conversely, if a kidney demonstrates half-time greater than 20 minutes and functional loss during early followup, urgent intervention is required for recurrent obstruction. REFERENCES
1. Chung, S., Majd, M., Rushton, H. G. et al: Diuretic renography in the evaluation of neonatal hydronephrosis: is it reliable? J Urol, 150: 765, 1993 2. Anderson, J. C.: Hydronephrosis: a fourteen year study of results. Proc R Soc Med, 55: 93, 1962 3. Notley, R. G. and Beaugie, J. M.: The long-term followup of Anderson-Hynes pyeloplasty for hydronephrosis. Br J Urol, 45: 464, 1973 4. Sunderland, H.: A review of experiences with the AndersonHynes plastic operation for hydronephrosis. Br J Urol, 35: 1, 1963 5. Roberts, M., Slade, N. and Jeffrey, P.: Late results in the management of primary pelvic hydronephrosis. Br J Urol, 44: 15, 1972 6. Williams, D. I. and Kenawai, M. M.: The prognosis of pelviureteric obstruction in childhood. Eur Urol, 2: 57, 1976 7. Stein, R., Ikoma, F., Salge, S. et al: Pyeloplasty in hydronephrosis: examination of surgical results from a morphologic point of view. Int J Urol, 3: 348, 1996
EARLY DIURESIS RENOGRAM FINDINGS PREDICT SUCCESS FOLLOWING PYELOPLASTY 8. Neste, M. G., Du Crete, R. P., Finlay, D. E. et al: Postoperative diuresis renography and ultrasound in patients undergoing pyeloplasty: predictors of surgical outcome. Clin Nucl Med, 18: 872, 1993 9. Fernbach, S. K., Maizels, M. and Conway, J. J.: Ultrasound grading of hydronephrosis: introduction to the system used by the Society for Fetal Urology. Pediatr Radiol, 23: 478, 1993 10. Amling, C. L., O’Hara, S. M., Wiener, J. S. et al: Renal ultrasound changes after pyeloplasty in children with ureteropelvic junction obstruction: long-term outcome in 47 renal units. J Urol, 156: 2020, 1996 11. Tapia, J. and Gonzalez, R.: Pyeloplasty improves renal function and somatic growth in children with ureteropelvic junction obstruction. J Urol, 154: 218, 1995
2315
12. Majd, M.: Nuclear medicine in pediatric urology. In: Clinical Pediatric Urology, 3rd ed. Edited by P. Kelalis, L. R. King and A. B. Belman. Philadelphia: W. B. Saunders Co., vol. 1, pp. 117–165, 1992 13. Butcher, H. R. and Sleator, W., Jr.: The effect of ureteral anastomosis upon conduction of peristaltic waves: an electroureterographic study. J Urol, 75: 650, 1956 14. Miyamoto, K. K. and Mesrobian, H. G.: Long-term outcome of kidneys with initially poor drainage or no drainage following pyeloplasty. World J Urol, 14: 380, 1996 15. Schlossberg, S. M.: Ureteral healing. Semin Urol, 5: 197, 1987 16. Salem, Y. H., Majd, M., Rushton, H. G. et al: Outcome analysis of pediatric pyeloplasty as a function of patient age, presentation, and differential renal function. J Urol, 154: 1889, 1995
Vol. 165, 2311–2315, June 2001 Printed in U.S.A.
EARLY DIURESIS RENOGRAM FINDINGS PREDICT SUCCESS FOLLOWING PYELOPLASTY HANS G. POHL, H. GIL RUSHTON, JAE-SHIN PARK, A. BARRY BELMAN
AND
MASSOUD MAJD
From the Departments of Pediatric Urology and Nuclear Medicine, Children’s National Medical Center, Washington, D. C.
ABSTRACT
Purpose: We sought to determine whether drainage across the ureteropelvic junction, as indicated by diuretic renography 3 months after pyeloplasty, is an adequate predictor of surgical success. Materials and Methods: The medical records of 150 children who underwent pyeloplasty from 1986 to 1995 were reviewed. After excluding nonevaluable cases a total of 127 renal units remained for investigation. Preoperatively each renal unit was examined with a standardized (well-tempered) furosemide stimulated renal scan. Postoperatively 60 renal units were evaluated with standardized diuretic renal scans at 3 and 12 months, 33 renal units at 3 months only and 34 renal units at 12 months only. Surgical success was defined by half-time less than 20 minutes on a standardized diuretic renogram. Results: Of the 33 renal units with a single postoperative study at 3 months 32 (97%) had halftime less than 20 minutes on diuretic renography. The remaining patient in this group with half-time greater than 20 minutes showed 60% improvement in half-time and did not require reoperation. Excluding those without delayed followup, surgical success was obtained in 93 of the 94 (99%) renal units. Among the 60 renal units evaluated with 2 postoperative renal scans success was noted in 48 (80%) and 59 (98%) at 3 and 12 months, respectively. Stenosis did not recur in 48 renal units with half-time less than 20 minutes 3 months after repair. In 1 case that had been treated for postoperative urinoma half-time was greater than 40 minutes at 3 months and repeat pyeloplasty was required. Conclusions: Half-time less than 20 minutes 3 months after pyeloplasty predicts surgical success. Most renal units that improve but still have half-times greater than 20 minutes on an early diuretic renogram will demonstrate continued improvement in drainage patterns at 12 months. Those renal units that show no improvement at 3 months may require reoperation and those with half-time less than 20 minutes at 3 months do not require further evaluation. KEY WORDS: diuresis, radioisotope renography, kidney
Diuretic renography remains the most objective means of assessing renal function and drainage associated with ureteropelvic junction obstruction,1 and a valid means of predicting postoperative success. It has been our impression that urinary drainage after successful repair of ureteropelvic junction obstruction has improved significantly by 3 months with little further improvement at 12 months. Therefore, we hypothesized that normalization of diuretic renography at 3 months postoperatively might obviate the need for long-term imaging at 1 year. Additionally, we propose that diuretic renography offers optimal assessment after pyeloplasty as a consequence of its ability to measure quantitatively relative renal function and urinary transit across the ureteropelvic junction obstruction. As a means of comparison we review reports of the use of excretory urography (IVP) and renal ultrasonography for post-pyeloplasty evaluation.
TABLE 1. Demographic data on 150 patients treated surgically for ureteropelvic junction obstruction % (No.) Gender: Males Females Race:* White Black Hispanic Asian Laterality:† Lt. Rt. Bilat. Pathology: Stenotic ureteropelvic junction obstruction Crossing vessel Fibroepithelial polyp * Race data not recorded in 11 patients. † Laterality not recorded in 8 patients.
MATERIALS AND METHODS
The medical records of 150 children (160 hydronephrotic renal units) who underwent pyeloplasty for ureteropelvic junction obstruction between 1986 and 1995 were reviewed. A total of 33 patients (33 units) were excluded from study because of insufficient renographic data (table 1). The remaining 117 patients (127 renal units) had undergone preoperative and postoperative standardized furosemide stimulated renal scans and divided into 1 of 3 groups according to the timing of postoperative followup (table 2). In group 1 postoperative scans were obtained at 3 and 12
65 (98) 35 (52) 71 (98) 19 (26) 7 (10) 3 (5) 55 (78) 38 (52) 7 (10) 87 (111) 12 (15) 1 (1)
months following repair in 60 cases. To determine whether the preoperative degree of obstruction influenced the rate of postoperative improvement in drainage half-times, the renal units in group 1 were further stratified using preoperative half-times of mild—20 to 40 minutes (23) moderate— 40 to 70 minutes (10) and severe— greater than 70 minutes (27). Group 2 consisted of 33 kidneys with complete furosemide renographic data only at 3 months after pyeloplasty. Of these
2311
2312
EARLY DIURESIS RENOGRAM FINDINGS PREDICT SUCCESS FOLLOWING PYELOPLASTY
TABLE 2. Surgical success following pyeloplasty for 127 renal units divided by the availability of postoperative renographic data Postop.
Group No.
Preop.
Early (3 mos.)
Late (12 mos.)
1 2 3
X X X
X X
X
% Success (No./total No.) 98 (59/60) 97 (32/33) 100 (34/34)
X
units 19 were lost to followup, repeat renography was done in 9 at 12 months but no half-time data were recorded and 5 units were evaluated with ultrasonography at 12 months rather than nuclear scintigraphy. Group 3 consisted of 34 cases with complete furosemide renographic data only at 12 months. For the purpose of this study, surgical success was defined as half-time less than 20 minutes postoperatively on a diuresis renogram. RESULTS
Of the group 1 renal units surgical success was demonstrated in 80% at 3 months and 98% at 12 months (table 3). Figure 1 illustrates the drainage patterns that typify the progressive improvement following successful pyeloplasty. When stratified as mild, moderate or severe ureteropelvic junction obstruction based on preoperative half-time values, improvement in half-time to less than 20 minutes at 3 months was noted in 90%, 80% and 70%, respectively (fig. 2). Those with postoperative half-time greater than 20 minutes demonstrated varying degrees of improvement in drainage compared with the preoperative renal scan, and relative function remained stable or improved (fig. 3). At 1 year improvement to half-time less than 20 minutes occurred in 100%, 100% and 96% (26 of 27), respectively, for cases of mild, moderate and severe obstruction, respectively (fig. 2). In 1 patient in the severe group a urinoma developed postoperatively and half-time was greater than 40 minutes at 3 months. At 12 months following initial repair worsening drainage evidenced by half-time greater than 100 minutes prompted a second pyeloplasty. Within 3 months of the repeat procedure half-time was 6 minutes. No patient with half-time less than 20 minutes at 3 months had late obstruction at 1 year. Of the 33 group 2 renal units 32 (97%) had half-time less than 20 minutes on diuretic renography. The patient with half-time greater than 20 minutes showed 60% improvement halftime and, to our knowledge, had no further difficulty (fig. 4). When groups 1 and 2 were combined 86% (80 of 93 units) of all renal units demonstrated half-time less than 20 minutes at 3 months postoperatively. Group 3 renal units had half-
FIG. 1. 99mTechnetium diethylenetetramine-pentoacetic acid diuresis renography of male infant presenting with antenatal hydronephrosis. A, preoperatively significant ureteropelvic junction obstruction is demonstrated by half-time greater than 100 minutes. B, 3 months after pyeloplasty half-time is significantly improved to less than 20 minutes. C, further improvement in drainage half-time is noted 12 months after surgery.
TABLE 3. Comparative results from previous reports of surgical outcome after Anderson-Hynes dismembered pyeloplasty with an emphasis on the method of postoperative followup % Improved at Interval From Surgery Method Followup
IVP: Williams and Kenawi6 Ultrasonography: Neste et al8 al10
No. Cases
1 Wk.
1 Mo.
3 Mos.
176
24 Mos.
Greater Than 24 Mos.
43† 61‡ 34
58† 74‡ 60 94
68† 84‡ 81
72† 96‡ 91
83
91
66.5*
46
47 8 Amling et 38 Tapia and Gonzalez11 14 37 81 92 Nephrostogram: Miyamoto and Mesrobian Renogram: 8 46 Neste et al 38 Tapia and Gonzalez11 Present study * Only minimal improvement in 46%. † SFU hydronephrosis grade used to assess postoperative improvement. ‡ Degree of renal pelvic dilatation used to assess postoperative improvement. § Includes all 93 renal units were evaluated at 3 months. 㛳 Includes all 94 renal units were evaluated at 12 months.
12 Mos.
6 Mos.
21
73 94 86§
99㛳
EARLY DIURESIS RENOGRAM FINDINGS PREDICT SUCCESS FOLLOWING PYELOPLASTY
FIG. 2. Success following pyeloplasty for 60 renal units evaluated with early and late postoperative diuretic renogram. Preoperative half-times were used to categorize renal units into 1 of 3 severity groups.
FIG. 3. Drainage patterns of 11 renal units evaluated by diuretic renography preoperatively and at 3 and 12 months postoperatively. This subset of group 2 includes all renal units that demonstrated half-times greater than 20 minutes 3 months after pyeloplasty. (Broken lines connect half-times of 3 instructive cases.) Of 11 renal units with early half-times greater than 20 minutes 9 (82%) demonstrated progressive improvement in drainage half-times to less than 20 minutes at 12 months (square). Additional renal unit improved to half-time 4 minutes 12 months after pyeloplasty, although diuretic renography at 3 months demonstrated persistently obstructed pattern (diamond). Remaining renal unit required repeat pyeloplasty to repair re-stenosis following urinoma (circle).
time less than 20 minutes on diuretic renography. Of all renal units from which renographic data were available 1 year postoperative (groups 1 and 3 combined) 99% (93 of 94 units) had half-time less than 20 minutes. When all 3 groups were considered 98.4% (125 of 127 units) had postoperative half-time less than 20 minutes at last followup. DISCUSSION
A number of methods have been applied to measure the postoperative results of pyeloplasty, including clinical status, IVP, ultrasonography and diuretic renography. Despite improvement in symptomatology following pyeloplasty, radiographic improvement in hydronephrosis has often been lacking.2– 4 Postoperative IVPs have routinely demonstrated
2313
FIG. 4. Preoperative and postoperative half-times from 30 group 2 renal units evaluated with diuretic renogram 3 months after repair. Solid line demonstrates 60% improvement in drainage pattern of only renal unit in group 2 with half-time greater than 20 minutes on “early” renogram. All remaining 29 renal units had half-times less than 20 minutes (hatched line). One case with half-time less than 20 minutes and biphasic time-activity curve consistent with significant flow dependent obstruction underwent dismembered pyeloplasty.
persistent caliceal clubbing and renal pelvic dilatation even on studies performed a decade after repair.3, 5 In fact, Roberts et al reported that caliceal clubbing improved in only 15 of 75 children (20%) after surgical correction.5 Similarly, in their review of 190 renal units treated with pyeloplasty Williams and Kenawi noted that calicectasis was unchanged or minimally improved in 79% of the cases.6 Only those renal units with minimal calicectasis preoperatively (10%) had significant improvement in caliceal clubbing on followup IVP. In general, when significant improvement was noted on IVP, it most commonly took the form of a reduction in renal pelvic dilatation while calicectasis persisted, which may have been a consequence of excision of a redundant renal pelvis. Stein et al reported similar findings in 193 renal units treated with dismembered pyeloplasty, of which 40 had grade 4 or “giant” hydronephrosis and a dilated renal pelvis that crossed the midline of the abdomen.7 In none of the renal units was the dilated pelvis excised in conjunction with resection of the stenotic ureteropelvic junction obstruction. Of the 40 renal pelves 24 (60%) improved to grade 1 or 2 on the last followup IVP. Despite symptomatic improvement in the remaining 40%, radiographic improvement was minimal or nonexistent (30% grade 3, 10% grade 4). These data suggest that the degree of pyelocalicectasis as demonstrated on IVP is an inaccurate measure of drainage. Additionally, the IVP affords only qualitative data on differential renal function. In 1993 Neste et al published post-pyeloplasty comparison between ultrasonography and diuretic renography of 56 renal units.8 Each unit was evaluated with ultrasonography and diuretic renography preoperatively and multiple times within the year following repair. Sonographic studies were graded according to the Society for Fetal Urology (SFU) classification criteria.9 Surgical success, defined as resolution of flank pain, absence of urinary infection and interval renal growth, occurred in 46 (82%) of the 56 renal units. The remaining 10 renal units associated with persistent flank pain, urinary infection and poor renal growth were deemed “unfavorable surgical outcomes” and were analyzed separately. Of the group that was clinically improved dilatation diminished in only 43% (12 of 28 units) of those evaluated with ultrasonography at 6 months (table 3). With longer intervals following surgery a greater percentage of renal units demonstrated diminished dilatation sonographically. By 24 months following pyeloplasty 72% (33 of 46 units) of those with clinical improvement demonstrated less dilatation. Among the 10 units with unfavorable outcomes 7 were unchanged and 2 demonstrated diminished hydronephrosis
2314
EARLY DIURESIS RENOGRAM FINDINGS PREDICT SUCCESS FOLLOWING PYELOPLASTY
on followup ultrasonography. Neste et al also investigated renal pelvic size alone, measured by greatest diameter in mm., as a predictor of surgical outcome. They reported initial improvement in 61% (17 of 28 units) at 6 months compared with 43% improvement noted when the SFU classification was used. Again, this difference is most likely secondary to surgical reduction of the renal pelvis. With longer followup to 18 months an additional 35% of the renal pelves demonstrated reduced diameters. Similarly, in a study of 44 children (47 kidneys) only 38% showed improvement on ultrasonography 6 months after pyeloplasty (table 3).10 Eventually, 81% demonstrated improvement according to sonographic criteria at 2 years. Thus, these data corroborate the finding of Stein et al that modest improvement in renal pelvic dilatation can be anticipated with time following successful pyeloplasty. Patient age at repair may have a role in the eventual resolution of pyelocalicectasis as noted on sonography or IVP. In the studies reported by Williams and Kenawi,6 and Stein et al,7 in which minimal improvement in calicectasis was noted, mean patient ages were 6.2 and 4.9 years, respectively. In contrast, in a younger cohort (mean age 2.5 years) Tapia and Gonzalez reported a reduction in the grade of hydronephrosis (by SFU classification) in 36 (94%) of the 38 renal units studied by ultrasonography 1 year after pyeloplasty (table 3).11 Despite significant hydronephrosis in all of the renal units treated (grade 3 in 10, grade 4 in 28), calicectasis eventually resolved completely in 75% of the kidneys. Since most ureteropelvic junction obstruction is now detected prenatally and repair is being performed at an earlier age, one might predict an increased number of renal units demonstrating sonographic improvement following pyeloplasty. However, sonographic data from Stein, Neste and Amling et al do not demonstrate that the infant kidney is more likely to have early resolution of pyelocalicectasis postoperatively.7, 8, 10 Since it is essential to detect persistent obstruction early to prevent irreversible loss of renal function, IVP and ultrasonography do not provide adequate assessment postoperatively. In contrast, diuretic renography offers a truly quantifiable assessment of renal function and drainage following repair that can be objectively compared with the preoperative scan. Consequently, postoperative obstruction can be detected promptly allowing for earlier intervention. The technique of diuretic renography has been standardized to minimize inter-examination variability. The standardized approach includes intravenous hydration with 15 ml./kg. dextrose 5%/0.3% normal saline beginning before tracer injection and continuing throughout the study, and placement of a bladder catheter.12 Furosemide (1 mg./kg., maximum 40 mg.) is administered only after maximal accumulation of the radiopharmaceutical has occurred in the dilated collecting system. Despite the commonly held misconception that the neonatal kidney is too immature to excrete radioisotopes and respond appropriately to furosemide challenge, diuretic renography has been shown to be reliable in the assessment of drainage and function in the infant kidney, even in patients as young as 3 days.1 Although it has been demonstrated that a period of healing is required between surgical repair and mucosal bridging (3 weeks), return of normal electrical conductivity (28 days) and completion of smooth muscle bridging (6 weeks), it has been demonstrated that the ureteropelvic junction obstruction can efficiently transmit a bolus of urine as early as 1 week following repair.13–15 Thus, it is most likely that improvement in drainage half-time is an early indicator of surgical outcome and precedes the morphological changes anticipated following relief of obstruction. In contemporary series success rates following pyeloplasty are reported to be between 72% and 98%.8, 10, 11, 14, 16 However, the results of postoperative evaluation vary with the
method of imaging study used and the timing of evaluation (table 3). Surgical success determined by diuretic renography appears to precede sonographic improvement by 9 months as demonstrated by the 73% to 94% success rate noted on renography at 3 months compared to 21% noted on ultrasonography (table 3). Not until 12 months have elapsed does the success rate as demonstrated on ultrasonography (58% to 94%) begin to approximate that observed on diuretic renography (91% to 98%). Our results concur with previously published findings of significant improvement in drainage by 3 months postoperatively, at which time 86% (80 of 93, groups 1 and 2 combined) of our patients had half-time less than 20 minutes.8, 11 We have also shown that the rate of early improvement following pyeloplasty correlates with the degree of preoperative obstruction based on diuretic renogram halftimes. Even so, the majority of kidneys (70%) with even severe obstruction preoperatively (half-time greater than 70 minutes) will by 3 months postoperatively demonstrate surgical success as defined by half-time less than 20 minutes. We acknowledge the possibility that stenosis could have recurred late in group 2 since delayed evaluation was not performed. However, it would be unlikely that this group alone had a significant proportion of late postoperative complications that was shielded from analysis. Furthermore, of those patients who were evaluated with diuretic renography at 3 and 12 months postoperatively none with half-time less than 20 minutes at 3 months demonstrated subsequent deterioration of drainage or late re-stenosis. In fact, 10 of 11 (91%) renal units with half-time greater than 20 minutes at 3 months postoperatively improved to half-time less than 20 minutes at 12 months. Therefore, the overall success rate for cases with followup at 1 year was 99% (93 of 94). The failure occurred in a patient with a postoperative urinoma who ultimately required repeat pyeloplasty for worsening drainage. One could argue that improvement on postoperative ultrasound at 3 months might obviate the need for a diuretic renogram. However, one would sacrifice the opportunity to evaluate postoperative function. Furthermore, since most sonograms are not performed under standardized hydration, the results could be misleading. CONCLUSIONS
When significant improvement is demonstrated on diuretic renography 3 months after dismembered pyeloplasty (halftime less than 20 minutes) no further followup is required. However, it does seem prudent to follow renal units with half-times greater than 20 minutes and no functional loss with additional diuretic renography performed 9 to 12 months after pyeloplasty. Conversely, if a kidney demonstrates half-time greater than 20 minutes and functional loss during early followup, urgent intervention is required for recurrent obstruction. REFERENCES
1. Chung, S., Majd, M., Rushton, H. G. et al: Diuretic renography in the evaluation of neonatal hydronephrosis: is it reliable? J Urol, 150: 765, 1993 2. Anderson, J. C.: Hydronephrosis: a fourteen year study of results. Proc R Soc Med, 55: 93, 1962 3. Notley, R. G. and Beaugie, J. M.: The long-term followup of Anderson-Hynes pyeloplasty for hydronephrosis. Br J Urol, 45: 464, 1973 4. Sunderland, H.: A review of experiences with the AndersonHynes plastic operation for hydronephrosis. Br J Urol, 35: 1, 1963 5. Roberts, M., Slade, N. and Jeffrey, P.: Late results in the management of primary pelvic hydronephrosis. Br J Urol, 44: 15, 1972 6. Williams, D. I. and Kenawai, M. M.: The prognosis of pelviureteric obstruction in childhood. Eur Urol, 2: 57, 1976 7. Stein, R., Ikoma, F., Salge, S. et al: Pyeloplasty in hydronephrosis: examination of surgical results from a morphologic point of view. Int J Urol, 3: 348, 1996
EARLY DIURESIS RENOGRAM FINDINGS PREDICT SUCCESS FOLLOWING PYELOPLASTY 8. Neste, M. G., Du Crete, R. P., Finlay, D. E. et al: Postoperative diuresis renography and ultrasound in patients undergoing pyeloplasty: predictors of surgical outcome. Clin Nucl Med, 18: 872, 1993 9. Fernbach, S. K., Maizels, M. and Conway, J. J.: Ultrasound grading of hydronephrosis: introduction to the system used by the Society for Fetal Urology. Pediatr Radiol, 23: 478, 1993 10. Amling, C. L., O’Hara, S. M., Wiener, J. S. et al: Renal ultrasound changes after pyeloplasty in children with ureteropelvic junction obstruction: long-term outcome in 47 renal units. J Urol, 156: 2020, 1996 11. Tapia, J. and Gonzalez, R.: Pyeloplasty improves renal function and somatic growth in children with ureteropelvic junction obstruction. J Urol, 154: 218, 1995
2315
12. Majd, M.: Nuclear medicine in pediatric urology. In: Clinical Pediatric Urology, 3rd ed. Edited by P. Kelalis, L. R. King and A. B. Belman. Philadelphia: W. B. Saunders Co., vol. 1, pp. 117–165, 1992 13. Butcher, H. R. and Sleator, W., Jr.: The effect of ureteral anastomosis upon conduction of peristaltic waves: an electroureterographic study. J Urol, 75: 650, 1956 14. Miyamoto, K. K. and Mesrobian, H. G.: Long-term outcome of kidneys with initially poor drainage or no drainage following pyeloplasty. World J Urol, 14: 380, 1996 15. Schlossberg, S. M.: Ureteral healing. Semin Urol, 5: 197, 1987 16. Salem, Y. H., Majd, M., Rushton, H. G. et al: Outcome analysis of pediatric pyeloplasty as a function of patient age, presentation, and differential renal function. J Urol, 154: 1889, 1995