THEJOURNAL OF UROLOGY Copyright O 1991 by AMERICANUROLOGICAL ASSOCIATION, INC
Vol. 145,6-10, January 1991 Printed in U.S.A.
Original Articles THE FATE OF RESIDUAL FRAGMENTS AFTER EXTRACORPOREAL SHOCK WAVE LITHOTRIPSY MONOTHERAPY OF INFECTION STONES EDWARD M. BECK
ROBERT A. RIEHLE, JR.
From the James Buchanan Brady Foundation, Department of Surgery, Division of Urology, T h e New York Hospital, Cornell Medical Center, New York, New York
We reviewed 53 patients with infection stones treated by extracorporeal shock wave lithotripsy (ESWL*) monotherapy to determine the long-term rate free of stones and the stone recurrence rate as correlated with the pre-treatment stone burden and the radiological presence of sand or fragments after the procedure. Long-term followup (mean 26.6 months) was available on 33 patients representing 38 kidneys. Although only 3 kidneys were free of stones immediately after ESWL, 20 were without stones a t 3 months and 18 (47%) were stone-free at followup. Of 9 kidneys with fragments of more than 5 mm. after the final treatment 7 (78%) had residual fragments at 3 months and experienced stone progression. Of 9 kidneys with sand remaining 6 (66%) and all 3 kidneys that appeared to be free of stones after ESWL were without stones at followup. The 3-month plain film of the kidneys, ureters and bladder was a reliable indicator of eventual outcome. Of 20 kidneys that were free of stones at 3 months 16 remained without stones. Of 18 kidneys with residual stone particles at 3 months 14 showed disease progression, 2 had stable disease and 2 passed residual sand. Only 1of 17 patients who were free of stones or had stable stone disease had a positive urine culture at followup. Patients with infection stone fragments 3 months after ESWL monotherapy have a high rate of stone progression (78%) and should undergo further treatment. ESWL monotherapy of infection stones requires close patient followup to assure that all residual fragments have passed and urine remains sterile. KEYWORDS:kidney calculi, lithotripsy
Urinary stones containing struvite account for approximately 15% of all calculi sent to the laboratory for analysis.' Modern crystallographic analysis has shown that infection stones are actually a mixture of struvite and carbonate apatite.' Historically, failure to treat this form of stone disease aggressively has led to significant morbidity and Extracorporeal shock wave lithotripsy (ESWL) has become the treatment of choice for most calculi of the upper urinary tract.6 However, infection stones (struvite and carbonate apatite) traditionally have been treated by open procedures and more recently by percutaneous nephrostolithotomy. Although ESWL monotherapy has proved to be effective in disintegrating struvite calculi, there h a s been concern about fragmenting infection stones into thousands of particles that may act as a nidus for stone regrowth a n d be a source of persistent infection. At our hospital many patients were discharged from the stone unit with radiological evidence of residual stone material for followup with the local urologists. We determined if these patients with kidneys containing infection stone particles passed the residual stone material or if retained sand and/or fragments led to recurrent infection or stone. In addition, we attempted to identify reliable indicators of eventual outcome. MATERIALS AND METHODS
Stone analyses of patients who had undergone ESWL monotherapy from July 1985 t o June 1987 were reviewed. Patients Accepted for publication June 20, 1990. * Dornier Medical Systems, Inc., Marietta, Georgia.
who underwent combined therapy with percutaneous nephrolithotomy and ESWL were not eligible. T o achieve a population of relatively pure stone composition only patients with stone analyses revealing greater than 80% struvite and carbonate apatite were considered. The patients were initially reviewed in a retrospective fashion. Pre-ESWL radiological studies and post-ESWL plain abdominal radiographs were reviewed by a single investigator (E. M. B.) without prior knowledge of eventual outcome. Stone burden was defined as the sum of the maximum diameter of all stones within 1 kidney. Stones were classified as solitary small (2 cm. or less), multiple small (2 cm. or less), solitary or multiple large (greater than 2 cm.), pelviocaliceal (renal pelvis extending into 1 calix), partial staghorn (renal pelvis extending into 2 calices) and complete staghorn (renal pelvis extending into 3 or more calices). Radiographs 1 day after the final treatment were classified as free of stones, sand remaining, fragments of less than 5 mm. remaining or fragments of greater than 5 mm. remaining. Followup evaluation was obtained by contacting each patient along with the initial referring urologist and the urologist at investigation. Radiographs performed a t our institution 3 months after ESWL were reviewed retrospectively. However, since most of the studies were performed by referring urologists, each physician reviewed the records and radiographs were classified as free of stones or with residual fragments remaining. When available, specific sizes of the residual fragments were obtained.
E X T R A C O R P O R E A L S K O C K WAVE L I T H O T R I P S Y E d O N B T H E R A P Y O F I N F E C T I O N S T O N E S
Patients were longitudinally followed by the local urologist or were offered followup at our institution. Any post-treatment report of cystitis, pyelonephritis, renal colic, stone passage or obstructive hydronephrosis requiring urological intervention was obtained. Followup examination included a physical examination, urine culture and a plain abdominal radiograph. The radiograph was examined for presence of stones as well as size of residual material. Stone progression, as defined by radiological criteria, was considered to be a 2-fold or greater increase in the diameter of residual fragments or recurrent stone formation after a state free of stones. Stable stone disease was considered to be no significant increase in the size of residual stone material remaining after ESWL. A total of 53 patients with 59 infection stone-containing kidneys was reviewed. Complete followup was available in 33 patients (62%) representing 38 kidneys. This group was representative of the over-all population (table 1). There were 30 women and 3 men with a mean age of 43 years (range 24 to 74 years). Fourteen patients underwent treatment of stones on the right side and 14 on the left side, while 5 had bilateral stones. All patients were neurologically intact and 2 had ileal conduit diversions. Mean followup was 26.6 months (range 11 to 40 months). Of 33 patients 19 (58%) were followed for 24 months or more: 10 underwent long-term followup a t our institution, while the remainder were evaluated by the local urologist. Urine culture results a t referral for ESWL yielded Proteus species in 16 patients, Klebsiella in 1, Pseudomonas in 1, Staphylococcus in 1, mixed flora in 5 and less than 100 colonies on antibiotics in 6. Culture results were not available in 3 patients. All patients underwent ESWL using the unmodified Dornier HM3 lithotriptor. Patients with pelviocaliceal stones had an average stone burden of 26 mm. Patients with partial and complete staghorn calculi had average stone burdens of 44 and 63 mm., respectively. All patients suspected of having infection stones, with or without positive urine cultures upon hospitalization, were placed on perioperative parenteral antibiotics and discharged from the hospital on oral antibiotics. Prolonged antimicrobial suppression was recommended until the patient was free of stones or displayed stable stone disease with negative urine cultures off antibiotics but patient compliance with this regimen was not complete. No patient underwent placement of a post-ESWL percutaneous nephrostomy tube for the purpose of renal irrigation. Urease inhibitors were not used.
Preoperative stone burden was not a reliable indicator of optimal (sand-like) disintegration as determined by postESWL or 3-month radiographs. Moreover, stone burden did not consistently correlate with an eventual state free of stone. Small solitary stones did not immediately fragment to sand in this series. However, 4 of these 5 stones were within a calix. Only 1 kidney with a caliceal stone was free of calculi at 3 months and this patient had a recurrent stone a t 18 months. One kidney had a stone in a dilated calix with thin overlying renal cortex and 1 kidney had a narrow infundibulum draining the stone-containing calix when previous excretory urograms were reviewed. The only kidney with a small solitary stone that was without calculi at followup had a renal pelvis calculus with sand on immediate post-ESWL and 3-month radiographs. A state free of stones was achieved in 50% of the kidneys with pelviocaliceal stones and 67% of those with partial staghorn calculi. However, no kidney with a complete staghorn calculus was free of stones upon followup. Post-ESWL fragment size as determined by a final postprocedure day 1 radiograph was a reliable indicator of eventual outcome only in several situations (table 2). All kidneys that were without stones after the final ESWL remained free of stones. Four kidneys with sand remaining after ESWL were free of stones at 3 months and a t followup. Two kidneys with sand after ESWL and residual stone material a t 3 months passed the residual fragments and were free of stones a t followup. Over-all, 6 of 9 kidneys (67%)with sand on radiography immediately after final ESWL were free of stones a t followup. Of 9 kidneys with fragments of greater than 5 mm. remaining after ESWL 7 (78%) had residual fragments at 3 months and stone progression on followup. In our series the 3-month post-treatment abdominal radiograph was the most reliable indicator of eventual outcome (table 3). Of 18 kidneys with residual fragments a t 3 months 14 (78%) had stone progression. However, in 2 kidneys residual fragments (sand) a t 3 months did pass and the kidneys were free of stones a t followup (32 and 15 months, respectively). Two kidneys in a patient with bilateral calculi have had stable stone disease during 35 months of followup. Of 20 kidneys that were free of stones at 3 months 16 (80%) remained so at followup and 4 had disease progression before long-term evaluation. A positive urine culture was uncommon in patients who were free of stones or showed no stone progression (table 4). Only 1 of 16 patients who were free of stones a t followup had a positive
TABLE2. Comparison of immediate post-ESWL radiograph results and long-term outcome
The 38 kidneys undergoing sequential monotherapy received an average of 1.3 treatments per kidney (range 1 to 2). Only 3 kidneys (8%) appeared to be free of stones when the patient was discharged from the hospital. However, 20 kidneys (53%) were free of stones a t 3 months and 18 (47%) were without stones a t followup. TABLE1. Profile of patients with infection stones Total (9%) No. pts. No. kidneys Mean age (yrs.) Male pts. Female pts. Pre-ESWL stone burden: Small solitary stone Multiple small stones Large stone ~elGiocalicea1stone Partial staghorn Complete staghorn Post-ESWL results: Free of stones Sand 5 mm. fragments
8 (13.5) 15 (25.5) 24 (41) 12 (20)
Free of stones Sand 5 mm. Totals
Followup Radiograph N ~with , 3 - ~ Long-Term ~ , Radiograph No, Free of No, With Kidneys Free of Stones Stones Progression NO. Stable 3 4 11 2 20
9 17 9 38
0 3 8 7 18
3 6 7 2 18
0 0 2 -0 2
Evaluable i % i 33 38 43 3 (9) 30 (91)
13 (22) 11 (18) 8 (14)
Immediate P,,~-ESWL Radiograph
5 (13) 3 (9) 8 (21) 9 (24) 8 (21) 5 (13) 3 (8) 9 (24) 17 (45) 9 (24)
TABLE 3. Fate of residual fragments at 3 months 3-Mo' Radiograph
Long-Term Followup Radiograph No. Kidneys No. Free of No. With Disease No. With Stones Progression Stable Disease
Residual fragments Free of stones
TABLE4. Long-term followup of urine cultures Stone Disease
Pos. Urine Culture
Free of stones Stable Progression
16 1 16
1 0 8
BECK AND RIEHLE
urine culture yielding Escherichia coli. One patient in this group had a previous urinary tract infection with Proteus species that responded to a short course of antibiotics and the urine has remained sterile since that time. One patient without stones had an ileal conduit and a urine culture was not sent by the referring urologist. The patient with stable stone disease had sterile urine a t the most recent followup although she was on chronic antibiotic suppression. Positive urine cultures were more common in patients who had experienced stone progression. Of 16 patients 8 had positive cultures on followup: Proteus species were isolated in 5 patients and E. coli in 3. In 1 patient with an ileal conduit a urine culture was not sent by the referring urologist. DISCUSSION
Infection stones are formed in urine infected with bacteria .~ hydrolyzes urea to that synthesize the enzyme ~ r e a s eUrease ammonia, bicarbonate and c a r b ~ n a t eThe . ~ subsequent alkaline urine allows crystallization of struvite and carbonate. The goal of treatment of infection stones has been the complete removal of all stone material and sterilization of urine. In the past the stones were routinely treated by pyelolithotomy and anatrophic nephrolithotomy. Open removal of infection and noninfection stones has resulted in a residual stone rate of 5 to 25%.*~',~ Recurrent stone rates have varied between 5 and 80%.3-5,8,9Griffith calculated the combined recurrence rate of infection stones from several surgical series to be approximately 30% a t 6 years.' Silverman and Stamey reported a residual stone rate of 24% and a recurrence rate of 2.2% in a series of patients with infection stones who received postoperative hemiacidrin renal irrigation.'' With refinements in less invasive techniques, infection stones of all sizes are currently being treated with percutaneous nephrolithotomy, ESWL or a combination of these 2 modalities. Initial ESWL investigations focused mainly on noiiiiifection type stones. This modality has been most effective in patients with solitary stones of less than 2 cm.11-13The residual stone rate was noted to increase as the number of stones or stone size increased. Although up to 85% of the patients will have radiological evidence of residual stones when they are discharged from the hospital, a t 3 months only 23 to 34% will have retained stone p a r t i ~ l e s . ~ l - l ~ Of great concern in treating infection stones with sequential ESWL monotherapy is t h e fate of the small fragments that remain when the patient is discharged from the hospital. This residual stone material could conceivably serve as a nidus of persistent infection and recurrent stone growth, especially in a distorted intrarenal collecting system. Winfield et al, comparing ESWL and percutaneous nephrolithotomy, found that almost all patients with staghorn calculi treated by ESWL monotherapy left the hospital with evidence of residual stones.14 However, a significant number of the patients passed the residual stone material and by 8 months 61% were free of stones. Since only a small number of patients with infection stones will be free of stones immediately after ESWL it is important to inquire which patients will have the best disintegration and, more importantly, which group of patients will pass the residual stone material and remain free of stones. Stone burden has traditionally been used as an indicator of future success when treating calculi with ESWL. However, our results suggest that stone burden in the setting of infection stones is not as reliable. I n our series only 20% of the kidneys with small solitary stones were free of calculi. A significant contributing factor in t h i s group was underlying abnormal intrarenal anatomy. However, even in kidneys with normal intrarenal anatomy stone burden before ESWL did not closely correlate with an eventual state free of stones. We found that a higher percentage of kidneys with partial staghorn calculi were free of stones on followup compared to those with pelvio-
caliceal stones. Of note, kidneys with multiple small calculi, large stones and partial staghorn calculi had similar stone-free rates. Schulze et a1 evaluated the role of combination therapy of infection stones and found that pre-ESWL stone burden alone did not account for the final treatment results.15 Since infection stones are soft and usually disintegrate well after ESWL, other factors besides stone size and composition are important. As with other forms of stone disease, stone location, intrarenal anatomy, body habitus and undertreatment can have a role. The crystallographic architecture of infection stones may also affect the pattern of disintegration. The most common size of spontaneously voided fragments after ESWL is 0.85 to 1.4 mm.16.17 However, Riehle et a1 demonstrated that uric acid and struvite/calcium carbonate mixtures (infection stones) tend to produce some of the largest fragments recovered (greater than 4 mm.).16 The large size of infection stone fragments often has been attributed to a large preoperative stone burden and, thus, less uniform and thorough disintegration a t the center of a staghorn calculus. Scanning electron microscopy has revealed that infection stones appear as an agglomeration of struvite and apatite crystals, cellular debris and other organic material.17 The distinct lack of homogeneity seen in infection stones also may account for the irregular disintegration patterns seen in our patients. Although the post-ESWL radiograph is often considered to be of limited value in assessing the degree of disintegration, we have found it to have prognostic significance if disintegration is clearly excellent (appears free of stones or sand remaining) or if fragmentation is limited. Of the kidneys containing what appeared to be fragments of greater than 5 mm. after the last sequential ESWL 78% experienced disease progression. If followed a t our institution it is likely that these patients would have undergone early retreatment. However, they were all followed by urologists outside our institution. In the majority of stone patients the degree of disintegration may not be apparent when they are discharged from the hospital and followup evaluation will be essential to plan future treatment. Proteus species account for the majority of infections seen in patients with infection s t o n e ~ . l ~However, -~~ voided urine cultures may not reflect the bacteriology of the upper tracts, and bacteria negative for urease may be incorporated into infection stones.'O Thus, there are several reasons why organisms negative for urease were identified in some of our patients who experienced disease progression. Our patients were not always compliant with the recommended use of post-ESWL antibiotics. Residual stone debris after ESWL may be susceptible to sterilization with routine use of antimicrobials. The systematic use of post-ESWL antibiotics in our patients may have resulted in a lower incidence of persistent infection and could have conceivably resulted in a lower rate of stone regrowth.'" " The importance of close followup in patients with infection stones cannot be over emphasized. Sixteen patients who we contacted by telephone had not been seen since the immediate post-ESWL period. In general, we recommend a followup abdominal radiograph and urine culture every 3 months until the patient is free of stones or demonstrates stable stone disease and the urine is sterile. Followup then is performed a t 6-month intervals. We believe that careful outpatient monitoring and an aggressive approach to the treatment of retained stones should be incorporated into the treatment plan for ESWL monotherapy of infection stones. More specifically, in patients with fragments larger than 5 mm. remaining on the radiograph 1 day after the last sequential ESWL a repeat radiograph should be performed within 2 to 3 weeks. If stone fragments are still present early retreatment or endourological intervention should be done. In patients who are free of stones or demonstrate sand on serial examinations early and close surveillance with voided
EXTRACORPOREAL SHOCK WAVE LITHOTRIPSY MONOTHERAPY OF INFECTION STONES cultures a n d serial a b d o m i n a l radiographs is indicated. S o m e p a t i e n t s w i t h s a n d r e m a i n i n g m a y eventually p a s s t h e residual s t o n e m a t e r i a l b u t m a y require longer t h a n 3 m o n t h s t o d o so. Finally, if followup e x a m i n a t i o n a t 3 m o n t h s fails t o reveal passage of f r a g m e n t s less t h a n 5 m m . o u r results support t h e s t a n c e that f u r t h e r t r e a t m e n t ( t h a t is dissolution t h e r a p y , ESWL, p e r c u t a n e o u s nephrolithotomy) is recommended. T h e s m a l l size and relatively large surface a r e a of residual f r a g m e n t s a f t e r ESWL should m a k e t h e s e p a t i e n t s ideal c a n didates for dissolution t h e r a p y . Indeed, initial reports of t h e m a n a g e m e n t of c o m p l e x s t r u v i t e stones w i t h ESWL a n d c h e m olysis w i t h h e m i a c i d r i n have b e e n e n c o ~ r a g i n g .However, ~~ p a t i e n t a c c e p t a n c e of an elective percutaneous nephrostomy t u b e for irrigation a n d dissolution of s m a l l particles a f t e r ESWL m a y b e limited.
18. Nemoy, N. J. and Stamey, T. A,: Surgical, bacteriological, and biochemical management of "infection stones". J.A.M.A., 215: 1470, 1971. 19. Rodman, J . S.: Struvite stones. In: Renal Stone Disease: Pathogenesis, Prevention and Treatment. Edited by C. Y. C. Pak. Boston: Martinus Nijhoff Publishing, chapt. 9, pp. 225-251, 1987. 20. Thompson, R. B. and Stamey, T. A,: Bacteriology of infected stones. Urology, 2: 627, 1973. 21. Fowler, J . E., Jr.: Bacteriology of branched renal calculi and accompanying urinary tract infection. J . Urol., 131: 213, 1984. 22. Michaels, E. K., Fowler, J . E. and Mariano, M.: Update on ESWL for infected renal stones. Infect. Urol., p. 33, March/April 1989. 23. Spirnak, J . P., DeBaz, B. P., Green, H . Y. and Resnick, M. I.: Complex struvite calculi treated by primary extracorporeal shock wave lithotripsy and chemolysis with hemiacidrin irrigation. J. Urol., 140: 1356, 1988.
ESWL m o n o t h e r a p y c a n be effective i n the t r e a t m e n t of infection s t o n e s b u t requires careful followup t o assure t h a t all s t o n e m a t e r i a l h a s p a s s e d a n d t h a t infection is n o t present. It is imperative t h a t t h e urologist t r e a t i n g infection s t o n e s h a s a complete a r s e n a l of medical a n d surgical modalities available a n d t h a t h e b e p e r s i s t e n t i n t r e a t i n g retained s t o n e f r a g m e n t s that d o n o t pass. REFERENCES
1. Williams, H. E.: Nephrolithiasis. New Engl. J . Med., 290: 33, 1974. 2. Griffith, D. P.: Struvite stones. Kidney Int., 13: 372, 1978. 3. Blandy, J . P. and Singh, M.: The case for a more aggressive approach to staghorn stones. J . Urol., 115: 505, 1976. 4. Vargas, A. D., Bragin, S. D. and Mendez, R.: Staghorn calculus: its clinical presentation, complications and management. J . Urol., 127: 860, 1982. 5. Rous, S. N. and Turner, W. R.: Retrospective study of 95 patients with staghorn calculus disease. J . Urol., 118: 902, 1977. 6. Riehle, R. A,, Jr., Naslund, E. B., Fair, W. and Vaughan, E. D., Jr.: Impact of shockwave lithotripsy on upper urinary tract calculi. Urology, 28: 261, 1986. 7. Griffith, D. P., Musher, D. M. and Itin, C.: Urease. The primary cause of infection-induced urinary stones. Invest. Urol., 13: 346, 1976. 8. Boyce, W. H. and Elkins, I. B.: Reconstructive renal surgery following anatrophic nephrolithotomy: followup of 100 consecutive cases. J . Urol., 11 1: 307, 1974. 9. Williams, R. E.: T h e results of conservative surgery for stone. Brit. J . Urol., 44: 292, 1972. 10. Silverman, D. E. and Stamey, T . A,: Management of infection stones: the Stanford experience. Medicine, 62: 44, 1983. 11. Riehle, R. A,, Jr., Fair, W. R. and Vaughan, E. D., Jr.: Extracorporeal shock-wave lithotripsy for upper urinary tract calculi: one year's experience at a single center. J.A.M.A., 255: 2043, 1986. 12. Drach, G. W., Dretler, S., Fair, W., Finlayson, B., Gillenwater, J., Griffith, D., Lingeman, J . and Newman, D.: Report of the United States cooperative study of extracorporeal shock wave lithotripsy. J . Urol., 135: 1127, 1986. 13. Lingeman, J. E., Newman, D., Mertz, J . H. O., Mosbaugh, P. G., Steele, R. E., Kahnoski, R. J . , Coury, T. A. and Woods, J . R.: Extracorporeal shock wave lithotripsy: the Methodist Hospital of Indiana experience. J. Urol., 135: 1134, 1986. 14. Winfield, H. N., Clayman, R. V., Chaussy, C. G., Weyman, P. J., Fuchs, G. J . and Lupu, A. N.: Monotherapy of staghorn renal calculi: a comparative study between percutaneous nephrolithotomy and extracorporeal shock wave lithotripsy. 3. Urol., 139: 895, 1988. 15. Schulze, H., Hertle, L., Kutta, A,, Graff, J . and Senge, T.: Critical evolution of treatment of staghorn calculi by percutaneous nephrolithotomy and extracorporeal shock wave lithotripsy. J . Urol., 141: 822, i989. 16. Riehle. R. A,., Jr.., Carter. H. B. and Vaughan. , E. D.. Jr.: Quantitative and crystallographic analysis of stone fragments koided after extracorporeal shock wave lithotripsy. J . Endourol., 1: 37, 1987. 17. Khan, S. R., Hackett, R. L. and Finlayson, B.: Morphology of urinary stone particles resulting from ESWL treatment. J. Urol., 136: 1367, 1986.
EDITORIAL COMMENTS The authors document the well established principle t h a t residual struvite stones increase in size with time. Of the kidneys t h a t became radiographically free of stones 80% remained so. Of the kidneys with residual stones 3 months after ESWL 78% showed stone growth. These data emphasize the need to educate patients about potential treatment alternatives and the need for long-term followup. These data also raise the question of whether long-term chemotherapy (with antimicrobials and/or urease inhibitors) should be used when there is residual debris. My practice is to advise patients about the risks of stone growth and the considerations involved in future operations, and the advantages and disadvantages of long-term chemotherapy. Use of medical treatment, thus, is tailored to the individual patient. Donald P. Griffith 5696 Longmont Houston, Texas This retrospective study clearly demonstrates the importance of complete pulverization of struvite renal calculi during ESWL. Of 29 stones that were fragmented to particles less than 5 mm. in diameter 18 (62%) were not identifiable or had residual nonexpansile particles at least 11 months after treatment. On the other hand, 7 of 9 stones (78%) that were fragmented to particles greater than 5 mm. in diameter were associated with particle enlargement. The latter probably is due to perpetuation of the persistent bacteriuria that is the cause of calculogenesis. Finely pulverized stone material can be sterilized with antibiotics' but bacteria within the interstices of large stones are refractory to this treatment. An issue that has not yet been clarified is whether residual sand or dust is a threat to the well-being of the patient. In this study 11 of the 26 stones (42%) that were pulverized to sand-like material had "progression." However, if I interpret the report correctly none of the affected patients was symptomatic or required subsequent treatments to remove the sand. Of the 16 patients with stone progression 8 and 1 with no identifiable stone fragments had a positive urine culture at the last followup. However, only 5 of the infections were caused by a ureasplitting bacterium and would predispose to new stone formation. Assuming that most or all of the patients with these isolates had residual particles more than 5 mm. in diameter the data suggest that the risk of persistent bacteriuria after proper stone fragmentation approaches zero. Our own experience with ESWL monotherapy for struvite renal calculi supports the concept that most small retained particles are of little clinical significance. We prospectively followed 22 women with documented persistent P . mirabilis bacteriuria before treatment for 16 to 48 months (mean 28 months). Antibiotics were discontinued 2 weeks after ESWL, or 2 weeks after the removal of ureteral stents or percutaneous nephrostomy tubes. Plain radiographs of the abdomen and urine cultures were obtained every 2 to 3 months. Of the patients 18 had radiographically identifiable clusters of fine stone particles within the collecting system after ESWL. In 12 cases the particles did not clear during 16 to 46 months (mean 26 months) of observation. However, none of the stones was symptomatic or increased in size. Three patients (2 with and 1 without retained particles) had persistent P. mirabilis bacteriuria during followup. Interestingly, 12 of the 19 women without persistent bacteriuria had urinary reinfections (infections caused by organisms other than P . mirabilis). One patient without residual particles maintained sterile urine for 20 months but subsequently had a P , mirabilis reinfection and a new stone in the previously