- Email: [email protected]
A PROSPECTIVE STUDY OF RECURRENCE RATE AND RISK FACTORS FOR RECURRENCE AFTER A FIRST RENAL STONE
A PROSPECTIVE STUDY O F RECURRENCE RATE AND RISK FACTORS FOR RECURRENCE AFTER A FIRST RENAL STONE ALBERT0 TRINCHIERI, F B I 0 OSTINI, ROBERTA NESPOLI, FABRIZIO ROVERA, EMANUELE MONTANARI AND GIAMPAOLO ZANETTI From the Department of Urology, Istituto Ricovero E Cura A Carattera ScientLfico, Ospedale Maggiore di Milano, Milan, Italy
ABSTRACT
Purpose: W e investigate f u r t h e r the recurrence rate and risk factors for recurrence in 300 consecutive p a t i e n t s who presented to o u r stone clinic after a first stone episode 7 to 17 y e a r s ago. Materials a n d Methods: The medical records of the patients who presented consecutively w i t h a first stone episode from 1980 t o 1990 were studied and supplemented by a followup mail questionnaire and telephone interviews. At first visit serum samples w e r e t a k e n from all patients a n d 24-hour urine samples w e r e collected for metabolic testing. Results: A total of 195 patients w e r e followed successfully, of whom 5 2 (27%) experienced symptomatic s t o n e recurrence after a mean plus or minus standard deviation of 7.5 5 5.9 years. However, ultrasound examination of 36 symptom-free patients showed recurrent stones in 28%. Comparison of patients w i t h or without recurrence confirmed that recurrence w a s not influenced by sex, family history of stones and u r i n a r y risk factors. However, age at onset of the disease was lower for p a t i e n t s w h o had 2 or more stones during followup than those who had only 1 stone or n o recurrence. Conclusions: S t o n e s c a n recur as long as 10 y e a r s after the first episode, although the rate is lower than previously reported. The metabolic evaluation after a first stone episode needs t o be reappraised in terms of its cost-effectiveness, since recurrences d o n o t seem to be predictable from standard laboratory tests. KEY WORDS:kidney calculi, recurrence, risk factors It is well known that kidney stones have a marked tendency to recur, with rates in some retrospective studies as high as 50% within 10 years of the first stone episode.1-6 However, these rates may be an overestimate since patients without new stones rarely return for visits and are often lost to followup. Although subsequent prospective study provided more reassuring figures,7 it remains difficult to establish which patients are likely to have a recurrent stone. There is widespread debate on the best clinical pathway t o take after a first stone episode as the type of diagnostic tests required must be weighed carefully and therapeutic measures must be aimed a t preventing further episodes. It is important to plan all necessary investigations t o ensure early etiological diagnosis but it can be costly to run through a long series of tests indiscriminately in all patients. With a view t o achieving an optimal cost-to-benefit ratio the risk factors for thorough evaluation need to be established. We assessed the likelihood of recurrence in light of various risk factors in a large population followed immediately after the first stone. MATERIALS AND METHODS
The series comprised 300 patients presenting consecutively with a first episode of renal stone from 1980 to 1990. Patients with remaining stones after the initial event were excluded from study as were those with residual fragments 3 months after endoscopic or extracorporeal lithotripsy. From our medical records we obtained patient sex and age, body weight and height, age at onset of symptoms, stone composition, related diseases (overt distal renal tubular acidosis, medullary sponge kidney, sarcoidosis, gout, urinary tract anomalies and other less frequent conditions) and modality of stone treatment. At visit 1 to the stone clinic, immediately after passage of
the stone or treatment, stone analysis, test for cystinuria and urine culture were performed. A plain film and ultrasound examination of the urinary tract were done to exclude false recurrence due to stones left behind. Serum from all patients was analyzed for potassium, sodium, calcium, phosphate, urate, creatinine and urea. A 24-hour urine sample was collected for determination of potassium, sodium, calcium, magnesium, phosphate, urate, oxalate, citrate, glycosaminoglycans, creatinine and urea. All patients were offered conservative treatment measures, including high fluid intake, and avoidance of excessive protein and salt intake. A low dairy food diet was never advised nor were agents such as thiazide, allopurinol or potassium citrate except for patients with cystine stones who received tiopronin. The subsequent history of urolithiasis was evaluated a t followup every 12 months, although many patients were lost to followup after several years. The medical records from 1980 to 1997 were studied. In April 1997 patients were mailed a questionnaire regarding the incidence and time of recurrence of new stones, passage of stones, extracorporeal, endoscopic or surgical stone treatment, or appearance of a stone on x-ray or ultrasound. A total of 165 patients completed the questionnaires. We attempted to interview the nonresponders by telephone and obtained 30 more replies. Two of the nonresponders were dead and 103 could not be contacted due to changes in address or new telephone numbers. A sample of 36 patients who stated on the questionnaire that they had no further stones were offered an ultrasound examination for confirmation. RESULTS
A total of 195 patients (65%)were available for mean followup of 13.9 t 5.7 years (minimum 7). Male-to-female ratio was 1:l and mean patient age was 44.3 ? 14.6 years
Accepted for publication January 29, 1999. 27
28
RECURRENCE AFTER FIRST RENAL STONE
(range 13 to 75). Patients who replied to the questionnaire and the nonresponders did not differ signrficantly in age, sex and stone characteristics. Stone composition was calcium in 147 cases (75.3%).infection in 29 (14.9%),uric acid or mixed uric acidkalcium oxalate in 17 (8.8%)and cystine in 2 (1%). At the initial investigation 2 patients had hyperparathyroidism which was treated surgically and 2 with cystine stones had cystinuria. Of the 195 patients 93 had passed the stone spontaneously and the remaining 102 underwent a procedure for stone removal. Nephrectomy was necessary in 9 cases, nephrolithotomy in 6, pyelolithotomy in 37 and ureterolithotomy in 11. The stone was removed by endoscopic percutaneous nephrolithotomy in 9 cases, ureteroscopy in 3 and extracorporeal shock wave lithotripsy in 19. In 7 patients the stone was treated with chemical dissolution or cystoscopic basket manipulation. Cases were grouped according to the date of the stone episode from 1980 to 1985 or 1986 to 1990 to assess the impact of new therapies on the course of disease. Overall, 52 patients experienced stone recurrence a h r a mean followup of 7.59 2 5.99 years (7.39 -C 6.03 for 28 men and 7.83 2 6.06 for 24 women). There were 19 patients with more than 1 recurrence but the mean interval between the first stone and first recurrence (5.21 2 4.99 years) was not significantly different from the mean interval between the first and subsequent stone recurrence (3.57 Z 3.64 years, p = 0.28). The rate of recurrence according to stone composition was 27% for calcium (40 of 147 cases), 20% for infection stones (6 of 29),23%for uric acid stones (3 of 131, and 50%for mixed (2 of 4) and cystine (1 of 2) stones. Survival analysis is shown in the figure. After 5 years 90% and after 10 years 78%of the patients had no clinical recurrence. During the first 2 years of followup approximately 3% of the patients at risk had recurrent stones each year. There-
R E
after and 10 years after the first attack 22%had at least 1 more stone. Further recurrences were reported in patients as long as 20 years after the first episode. Ultrasound examination of the 36 symptom-free patients who stated that they had had no recurrences showed another stone in 5 and microlithiasis (3 mm. or smaller) in 5. There was a difference, although not significant (chisquare 0.07), in recurrence rates between the 165 patients who returned the questionnaire and the 30 contacted by telephone (29 versus 13%).Using chi-square we found that male sex and family history were not significant risk factors for recurrence (table 1). However, age at stone onset was useful for identifying patients a t particular risk for 2 or more stones. In calcium stone formers serum and urinary parameters were not useful as predictors of subsequent recurrence, except that urinary pH was higher in patients with than in those without recurrences (table 2). Likewise, the incidence of metabolic defects was similar in patients with and without recurrences. The incidence in these 2 groups was 27 and 26.1% for hypercalciuria, 8.6 and 14.3% for hyperuricuria, 27.2 and 18.1% for hyperoxaluria, and 28.6 and 27.1% for hypocitruria, respectively. The rate of recurrence was 24% aRer spontaneous passage (23 of 93 cases), 26% after open surgery (17 of 63), 30% after endoscopic procedure (4 of 13) and 21% after extracorporeal lithotripsy (4 of 19). No difference in recurrence rate at 5 years was observed between patients treated from 1980 to 1985 (14 of 103,13%)and those treated from 1986 to 1990 (14 of 92, 15%). DISCUSSION
Until the 1950s it was believed that many stone episodes remained isolated and linked to chance events that were unlikely to recur. However, subsequent studies using more
I
R R
E
E
A
YEARS
YEARS
B ..I
R E
c
U
.*
R R L.....
L
E
'6
E
.7.
R A T E
.5
N C
!.........................
i l , -
. uc
' U -
,
. ST ...... '
CY
__-
j 'C 10
C
YEARS
j -:; -
.a
20
YEARS
D
Recurrence-free rate after first renal stone calculated with Cox's regression method (A), in relation to sex ( B ) ,stone composition (C) and a e at stone onset (D). M, male patients. F, female atients. C , calcium. U,uric acid. UC, mixed uric acidcalcium. ST,struvite. CY, cystine.
RECURRENCE AFTER FIRST RENAL STONE
1. Risk factors for recurrence
29
with followup was so low that afterwards we had to mail a questionnaire to update our information retrospectively. ~_ _ _ . ~Significance ~ However, our design is more similar to the previously cited 0 1 More than 1 prospective than to retrospective studies. Our incidence of No. pts. 143 33 19 recurrence was close to that of Ahlstrand and Tiselius.7 and No. nien/No. women 70173 16/17 12/7 0.44 lower than that of Ljunghall and Danielson,n as well as that No. family history ( % ) 35 (24.5) 9 127.2) 2 (10.51 0.33 of retrospective studies. There were several sources of bias Mean age at stone 44 t- 14 44 t 11 35 2 13 0.036 onset 2 SD that led to our underestimated incidence of recurrence. The fact that x-rays and ultrasound examinations were not required systematically during followup could have resulted in TABLE2. Urinary risk factors for recurrence in calcium renal stone oversight of symptom-free cases. The percentage of replies to formers the questionnaire was fairly low, and the male-to-female ratio was not typical of the normal renal stone population. Mean -C SD 1 More than 1 Significance However, the long followup should limit the proportion of symptom-free patients, and the clinical significance of any O Recurrence Recurrence Recurrence unnoticed recurrences is obviously slight if patients had no Vol. (ml./day) 1,617f 653 1,364 f 569 1,584 2 500 0.27 clinical signs for such a long period. Ultrasound examination Calcium (mgJday) 220 t 113 251 f 128 0.47 249 2 97 of a sample of patients whose questionnaire replies indicated Urate (mg./day) 679 % 668 479 f 155 0.35 511 2 216 Oxalate (rng./day) 28 t- 22 28 f 9 0.50 42 2 31 they had no recurrences confirmed that about 28%, of them Citrate (rng./day) 503 t 194 338 f 126 450 2 50 0.10 did in fact have asymptomatic stones, although more than 5.73 t- 0.61 5.93 f 0.51 6.23 2 0.59 DH 0.018 half of these were clinically insignificant (less than 3 mm.). The low percentage of replies to the questionnaire appeared to be due not so much to a lack of interest in the rigorous methods and longer observation periods revealed investigation as to the fact that so many patients could not be that stones recurred a long time after the first episode. Wil- contacted due to change in address or telephone number. liams reported recurrences in 75% of patients observed for However, comparison of the incidence of recurrences indimore than 10 years (mean followup 18.5 years) after the first cated on the questionnaire with the proportion from our stone episode.’ Marshall et a1 compared 40.3% of recurrences telephone interviews indicated that nonresponders subsein men with 30.1% in women, and noted that it was hard to quently reached by telephone who had not had recurrences predict recurrences but that the risk persisted for a long time were less interested in taking part in the study. Thus, the and declined slowly within years.2 Ljunghall et a1 reported questionnaire probably did not give an underestimate and similar results.3 Replies to a postal questionnaire indicated may even have achieved the opposite effect. that 46% of men and 43% of women had a recurrence after a There are several explanations why half of the patients first stone episode during an observation period of 15 to 19 were women. The percentage of the replies to the questionyears. A health screening survey confirmed that 45% of patients were likely to have recurrence within 10 years of the naire was lower in male (98of 156,63%) than in female (97 first stone.4 In a rural population in the United States Suth- of 144,67%) patients who appeared to be more interested in erland et a1 reported 50% of recurrences 8.8 years after the taking part in the study. About 15%of our patients presented first stone episode,5 while Johnson et al noted 40% of recur- with infection stones that were less common in men (4) than rences 10 years after the first attack.6 These retrospective in women (25).Our patients represent an unselected popustudies led to the belief that there was a high risk of recur- lation with less severe disease. On the other hand, others rence after a first stone episode and, consequently, that all have found gender to be almost equal among unselected patients required a thorough metabolic evaluation and meas- stone forming patients.” However, in our study the rate of recurrence was similar (29% for men and 24% for women). ures to prevent recurrences. Studies of risk factors to identify patients most likely to In 2 prospective studies Ljunghall and Danielson confirmed that 50% of their patients had recurrence within 8 have recurrence often give disappointing results. In general years of the first episode8 but Ahlstrand and Tiselius re- women tend to have recurrence less often than men,1.2.5-R ported that only 26% of their patients had recurrence within although this has not always been confirmed.s.7 Family his10 years after the first stone.7 The different results may be tory of renal stones is another risk factor in some7.8 but not due to differences in how the studies were conducted, study all5 studies. Sutherland et a1 did not find that young age at inclusion criteria, mean patient age, male-to-female ratio, onset was a risk factor. Only Ljunghall and Danielson conlength of followup, diagnostic methods, followup criteria and sidered high uric acid to be a risk factor.8 Others have prepreventive measures. Differences between retrospective and viously demonstrated that first time and recurrent stone prospective studies are important. Retrospective studies tend formers have similar metabolic profiles.12 However, many to overestimate the incidence of recurrence as they are likely studies did not include complete metabolic evaluation for all to select a certain population recruited from specialized cen- patients and retrospective studies made the biochemical asters a t which patients with recurring stones are primarily sessment several years after the first stone episode. There treated rather than first stone cases. This bias does not apply have been reports of Proteus infection in women (significant obviously to investigations from outlying or isolated centers,s for infection stones)* and multiple stones13 as risk factors for or those conducted using questionnaires mailed to unselected populations.3~4The objection in the latter case might be that recurrence, which indicates that some studies did not distinpatients with recurrent stones may be more motivated than guish clearly between the different types of stone. Some others to answer a questionnaire. However, prospective stud- studies included patients receiving pharmacological treaties may underestimate the likelihood of recurrence when ment to prevent recurrences, which may have influenced the they do not specify systematic x-ray or ultrasound to detect course of disease.2.7 We found no real differences between asymptomatic stones with an incidence of about 30%.9.10A patients with and without recurrence in regard to male-tureason for the differences in the 2 aforementioned prospec- female ratio, family history of stones, age at onset or incitive studies is the different proportion of women (33%7 and dence of various metabolic disorders. However, we did note that patients with 1 versus more than 1 recurrence were 20%8). The design of our study was prospective but compliance younger at onset of disease. TABLE
Recurrence
~
30
RECURRENCE AFTER FIRST RENAL STONE CONCLUSIONS
Our findings indicate that stones can recur many years after t h e first episode, although our percentage of recurrences was lower t h a n t h at in earlier reports. As even a complete metabolic evaluation including 24-hour urinalysis did not identify patients at risk for recurrence after the first stone episode, it might be worthwhile to evaluate secondary disorders, such as primary hyperparathyroidism and cystinuria. Male patients require close followup as first stone occurred when they were young. The new minimally invasive treatments did not appear to have any appreciable impact on recurrence since in 7 years of followup we found no differences in the incidence of recurrent stones among patients treated in the 5 years before shock wave lithotripsy and those treated after we began to use the less invasive techniques. REFERENCES
1. Williams, R. E.: Long-term survey of 538 patients with upper urinary tract stone. Brit. J. Urol., 3 5 416, 1963. 2. Marshall, V., White, R. H., De Saintonge, M. C., Tresidder, G. C. and Blandy, J. P.: The natural history of renal and ureteric calculi. Brit. J . Urol., 47: 117, 1975. 3. Ljunghall, S. and Hedstrand, H.: Epidemiology of renal stones in a middle-aged male population. Acta Med. Scand., 97: 439, 1975.
4. Ljunghall, S.: Incidence and natural history of renal stone disease and its relationship to calcium metabolism. Eur. Urol., 4: 424, 1978. 5. Sutherland. J . W., Parks, J. H. and Coe, F. L.: Recurrence after a single renal stone in a community practice. Miner. Electrolyte Metab., 11: 267, 1985. 6. Johnson, C. M., Wilson, D. M., OFallon, W. M., Malek, R. S. and Kurland, L. T.: Renal stone epidemiology: a 25-year study in Rochester, Minnesota. Kidney Int., 1 6 624, 1979. 7. Ahlstrand, C. and Tiselius, H. G.: Recurrences during a 10-year follow-up after first renal stone episode. Urol. Res., 18: 397, 1990. 8. Ljunghall. S., Danielson, B. G.: A prospective study of renal stone recurrences. Brit. J. Urol., 56: 122, 1984. 9. Hellstrom, J.: Aetiological and therapeutic experiences concerning kidney and ureteric stones. Brit. J. Urol., 21: 9, 1949. 10. Sutherland, J. W.: Recurrence following operative treatment of upper urinary tract stone. J. Urol., 127: 472, 1982. 11. Scott, R.: Prevalence of calcified upper urinary tract stone disease in a random population survey. Report of a combined study of general practitioners and hospital staff. Brit. J. Urol., 5 9 111, 1987. 12. Pak, C. Y. C . : Should patients with single renal stone occurrence undergo diagnostic evaluation? J. Urol., 127: 855, 1982. 13. Sun, B. Y.-C., Lee, Y.-H., Jiaan, B.-P., Chen, K.-K., Chang, L. S. and Chen, K.-T.: Recurrence rate and risk factors for urinary calculi after extracorporeal shock wave lithotripsy. J. Urol., 156.903, 1996.
ABSTRACT
Purpose: W e investigate f u r t h e r the recurrence rate and risk factors for recurrence in 300 consecutive p a t i e n t s who presented to o u r stone clinic after a first stone episode 7 to 17 y e a r s ago. Materials a n d Methods: The medical records of the patients who presented consecutively w i t h a first stone episode from 1980 t o 1990 were studied and supplemented by a followup mail questionnaire and telephone interviews. At first visit serum samples w e r e t a k e n from all patients a n d 24-hour urine samples w e r e collected for metabolic testing. Results: A total of 195 patients w e r e followed successfully, of whom 5 2 (27%) experienced symptomatic s t o n e recurrence after a mean plus or minus standard deviation of 7.5 5 5.9 years. However, ultrasound examination of 36 symptom-free patients showed recurrent stones in 28%. Comparison of patients w i t h or without recurrence confirmed that recurrence w a s not influenced by sex, family history of stones and u r i n a r y risk factors. However, age at onset of the disease was lower for p a t i e n t s w h o had 2 or more stones during followup than those who had only 1 stone or n o recurrence. Conclusions: S t o n e s c a n recur as long as 10 y e a r s after the first episode, although the rate is lower than previously reported. The metabolic evaluation after a first stone episode needs t o be reappraised in terms of its cost-effectiveness, since recurrences d o n o t seem to be predictable from standard laboratory tests. KEY WORDS:kidney calculi, recurrence, risk factors It is well known that kidney stones have a marked tendency to recur, with rates in some retrospective studies as high as 50% within 10 years of the first stone episode.1-6 However, these rates may be an overestimate since patients without new stones rarely return for visits and are often lost to followup. Although subsequent prospective study provided more reassuring figures,7 it remains difficult to establish which patients are likely to have a recurrent stone. There is widespread debate on the best clinical pathway t o take after a first stone episode as the type of diagnostic tests required must be weighed carefully and therapeutic measures must be aimed a t preventing further episodes. It is important to plan all necessary investigations t o ensure early etiological diagnosis but it can be costly to run through a long series of tests indiscriminately in all patients. With a view t o achieving an optimal cost-to-benefit ratio the risk factors for thorough evaluation need to be established. We assessed the likelihood of recurrence in light of various risk factors in a large population followed immediately after the first stone. MATERIALS AND METHODS
The series comprised 300 patients presenting consecutively with a first episode of renal stone from 1980 to 1990. Patients with remaining stones after the initial event were excluded from study as were those with residual fragments 3 months after endoscopic or extracorporeal lithotripsy. From our medical records we obtained patient sex and age, body weight and height, age at onset of symptoms, stone composition, related diseases (overt distal renal tubular acidosis, medullary sponge kidney, sarcoidosis, gout, urinary tract anomalies and other less frequent conditions) and modality of stone treatment. At visit 1 to the stone clinic, immediately after passage of
the stone or treatment, stone analysis, test for cystinuria and urine culture were performed. A plain film and ultrasound examination of the urinary tract were done to exclude false recurrence due to stones left behind. Serum from all patients was analyzed for potassium, sodium, calcium, phosphate, urate, creatinine and urea. A 24-hour urine sample was collected for determination of potassium, sodium, calcium, magnesium, phosphate, urate, oxalate, citrate, glycosaminoglycans, creatinine and urea. All patients were offered conservative treatment measures, including high fluid intake, and avoidance of excessive protein and salt intake. A low dairy food diet was never advised nor were agents such as thiazide, allopurinol or potassium citrate except for patients with cystine stones who received tiopronin. The subsequent history of urolithiasis was evaluated a t followup every 12 months, although many patients were lost to followup after several years. The medical records from 1980 to 1997 were studied. In April 1997 patients were mailed a questionnaire regarding the incidence and time of recurrence of new stones, passage of stones, extracorporeal, endoscopic or surgical stone treatment, or appearance of a stone on x-ray or ultrasound. A total of 165 patients completed the questionnaires. We attempted to interview the nonresponders by telephone and obtained 30 more replies. Two of the nonresponders were dead and 103 could not be contacted due to changes in address or new telephone numbers. A sample of 36 patients who stated on the questionnaire that they had no further stones were offered an ultrasound examination for confirmation. RESULTS
A total of 195 patients (65%)were available for mean followup of 13.9 t 5.7 years (minimum 7). Male-to-female ratio was 1:l and mean patient age was 44.3 ? 14.6 years
Accepted for publication January 29, 1999. 27
28
RECURRENCE AFTER FIRST RENAL STONE
(range 13 to 75). Patients who replied to the questionnaire and the nonresponders did not differ signrficantly in age, sex and stone characteristics. Stone composition was calcium in 147 cases (75.3%).infection in 29 (14.9%),uric acid or mixed uric acidkalcium oxalate in 17 (8.8%)and cystine in 2 (1%). At the initial investigation 2 patients had hyperparathyroidism which was treated surgically and 2 with cystine stones had cystinuria. Of the 195 patients 93 had passed the stone spontaneously and the remaining 102 underwent a procedure for stone removal. Nephrectomy was necessary in 9 cases, nephrolithotomy in 6, pyelolithotomy in 37 and ureterolithotomy in 11. The stone was removed by endoscopic percutaneous nephrolithotomy in 9 cases, ureteroscopy in 3 and extracorporeal shock wave lithotripsy in 19. In 7 patients the stone was treated with chemical dissolution or cystoscopic basket manipulation. Cases were grouped according to the date of the stone episode from 1980 to 1985 or 1986 to 1990 to assess the impact of new therapies on the course of disease. Overall, 52 patients experienced stone recurrence a h r a mean followup of 7.59 2 5.99 years (7.39 -C 6.03 for 28 men and 7.83 2 6.06 for 24 women). There were 19 patients with more than 1 recurrence but the mean interval between the first stone and first recurrence (5.21 2 4.99 years) was not significantly different from the mean interval between the first and subsequent stone recurrence (3.57 Z 3.64 years, p = 0.28). The rate of recurrence according to stone composition was 27% for calcium (40 of 147 cases), 20% for infection stones (6 of 29),23%for uric acid stones (3 of 131, and 50%for mixed (2 of 4) and cystine (1 of 2) stones. Survival analysis is shown in the figure. After 5 years 90% and after 10 years 78%of the patients had no clinical recurrence. During the first 2 years of followup approximately 3% of the patients at risk had recurrent stones each year. There-
R E
after and 10 years after the first attack 22%had at least 1 more stone. Further recurrences were reported in patients as long as 20 years after the first episode. Ultrasound examination of the 36 symptom-free patients who stated that they had had no recurrences showed another stone in 5 and microlithiasis (3 mm. or smaller) in 5. There was a difference, although not significant (chisquare 0.07), in recurrence rates between the 165 patients who returned the questionnaire and the 30 contacted by telephone (29 versus 13%).Using chi-square we found that male sex and family history were not significant risk factors for recurrence (table 1). However, age at stone onset was useful for identifying patients a t particular risk for 2 or more stones. In calcium stone formers serum and urinary parameters were not useful as predictors of subsequent recurrence, except that urinary pH was higher in patients with than in those without recurrences (table 2). Likewise, the incidence of metabolic defects was similar in patients with and without recurrences. The incidence in these 2 groups was 27 and 26.1% for hypercalciuria, 8.6 and 14.3% for hyperuricuria, 27.2 and 18.1% for hyperoxaluria, and 28.6 and 27.1% for hypocitruria, respectively. The rate of recurrence was 24% aRer spontaneous passage (23 of 93 cases), 26% after open surgery (17 of 63), 30% after endoscopic procedure (4 of 13) and 21% after extracorporeal lithotripsy (4 of 19). No difference in recurrence rate at 5 years was observed between patients treated from 1980 to 1985 (14 of 103,13%)and those treated from 1986 to 1990 (14 of 92, 15%). DISCUSSION
Until the 1950s it was believed that many stone episodes remained isolated and linked to chance events that were unlikely to recur. However, subsequent studies using more
I
R R
E
E
A
YEARS
YEARS
B ..I
R E
c
U
.*
R R L.....
L
E
'6
E
.7.
R A T E
.5
N C
!.........................
i l , -
. uc
' U -
,
. ST ...... '
CY
__-
j 'C 10
C
YEARS
j -:; -
.a
20
YEARS
D
Recurrence-free rate after first renal stone calculated with Cox's regression method (A), in relation to sex ( B ) ,stone composition (C) and a e at stone onset (D). M, male patients. F, female atients. C , calcium. U,uric acid. UC, mixed uric acidcalcium. ST,struvite. CY, cystine.
RECURRENCE AFTER FIRST RENAL STONE
1. Risk factors for recurrence
29
with followup was so low that afterwards we had to mail a questionnaire to update our information retrospectively. ~_ _ _ . ~Significance ~ However, our design is more similar to the previously cited 0 1 More than 1 prospective than to retrospective studies. Our incidence of No. pts. 143 33 19 recurrence was close to that of Ahlstrand and Tiselius.7 and No. nien/No. women 70173 16/17 12/7 0.44 lower than that of Ljunghall and Danielson,n as well as that No. family history ( % ) 35 (24.5) 9 127.2) 2 (10.51 0.33 of retrospective studies. There were several sources of bias Mean age at stone 44 t- 14 44 t 11 35 2 13 0.036 onset 2 SD that led to our underestimated incidence of recurrence. The fact that x-rays and ultrasound examinations were not required systematically during followup could have resulted in TABLE2. Urinary risk factors for recurrence in calcium renal stone oversight of symptom-free cases. The percentage of replies to formers the questionnaire was fairly low, and the male-to-female ratio was not typical of the normal renal stone population. Mean -C SD 1 More than 1 Significance However, the long followup should limit the proportion of symptom-free patients, and the clinical significance of any O Recurrence Recurrence Recurrence unnoticed recurrences is obviously slight if patients had no Vol. (ml./day) 1,617f 653 1,364 f 569 1,584 2 500 0.27 clinical signs for such a long period. Ultrasound examination Calcium (mgJday) 220 t 113 251 f 128 0.47 249 2 97 of a sample of patients whose questionnaire replies indicated Urate (mg./day) 679 % 668 479 f 155 0.35 511 2 216 Oxalate (rng./day) 28 t- 22 28 f 9 0.50 42 2 31 they had no recurrences confirmed that about 28%, of them Citrate (rng./day) 503 t 194 338 f 126 450 2 50 0.10 did in fact have asymptomatic stones, although more than 5.73 t- 0.61 5.93 f 0.51 6.23 2 0.59 DH 0.018 half of these were clinically insignificant (less than 3 mm.). The low percentage of replies to the questionnaire appeared to be due not so much to a lack of interest in the rigorous methods and longer observation periods revealed investigation as to the fact that so many patients could not be that stones recurred a long time after the first episode. Wil- contacted due to change in address or telephone number. liams reported recurrences in 75% of patients observed for However, comparison of the incidence of recurrences indimore than 10 years (mean followup 18.5 years) after the first cated on the questionnaire with the proportion from our stone episode.’ Marshall et a1 compared 40.3% of recurrences telephone interviews indicated that nonresponders subsein men with 30.1% in women, and noted that it was hard to quently reached by telephone who had not had recurrences predict recurrences but that the risk persisted for a long time were less interested in taking part in the study. Thus, the and declined slowly within years.2 Ljunghall et a1 reported questionnaire probably did not give an underestimate and similar results.3 Replies to a postal questionnaire indicated may even have achieved the opposite effect. that 46% of men and 43% of women had a recurrence after a There are several explanations why half of the patients first stone episode during an observation period of 15 to 19 were women. The percentage of the replies to the questionyears. A health screening survey confirmed that 45% of patients were likely to have recurrence within 10 years of the naire was lower in male (98of 156,63%) than in female (97 first stone.4 In a rural population in the United States Suth- of 144,67%) patients who appeared to be more interested in erland et a1 reported 50% of recurrences 8.8 years after the taking part in the study. About 15%of our patients presented first stone episode,5 while Johnson et al noted 40% of recur- with infection stones that were less common in men (4) than rences 10 years after the first attack.6 These retrospective in women (25).Our patients represent an unselected popustudies led to the belief that there was a high risk of recur- lation with less severe disease. On the other hand, others rence after a first stone episode and, consequently, that all have found gender to be almost equal among unselected patients required a thorough metabolic evaluation and meas- stone forming patients.” However, in our study the rate of recurrence was similar (29% for men and 24% for women). ures to prevent recurrences. Studies of risk factors to identify patients most likely to In 2 prospective studies Ljunghall and Danielson confirmed that 50% of their patients had recurrence within 8 have recurrence often give disappointing results. In general years of the first episode8 but Ahlstrand and Tiselius re- women tend to have recurrence less often than men,1.2.5-R ported that only 26% of their patients had recurrence within although this has not always been confirmed.s.7 Family his10 years after the first stone.7 The different results may be tory of renal stones is another risk factor in some7.8 but not due to differences in how the studies were conducted, study all5 studies. Sutherland et a1 did not find that young age at inclusion criteria, mean patient age, male-to-female ratio, onset was a risk factor. Only Ljunghall and Danielson conlength of followup, diagnostic methods, followup criteria and sidered high uric acid to be a risk factor.8 Others have prepreventive measures. Differences between retrospective and viously demonstrated that first time and recurrent stone prospective studies are important. Retrospective studies tend formers have similar metabolic profiles.12 However, many to overestimate the incidence of recurrence as they are likely studies did not include complete metabolic evaluation for all to select a certain population recruited from specialized cen- patients and retrospective studies made the biochemical asters a t which patients with recurring stones are primarily sessment several years after the first stone episode. There treated rather than first stone cases. This bias does not apply have been reports of Proteus infection in women (significant obviously to investigations from outlying or isolated centers,s for infection stones)* and multiple stones13 as risk factors for or those conducted using questionnaires mailed to unselected populations.3~4The objection in the latter case might be that recurrence, which indicates that some studies did not distinpatients with recurrent stones may be more motivated than guish clearly between the different types of stone. Some others to answer a questionnaire. However, prospective stud- studies included patients receiving pharmacological treaties may underestimate the likelihood of recurrence when ment to prevent recurrences, which may have influenced the they do not specify systematic x-ray or ultrasound to detect course of disease.2.7 We found no real differences between asymptomatic stones with an incidence of about 30%.9.10A patients with and without recurrence in regard to male-tureason for the differences in the 2 aforementioned prospec- female ratio, family history of stones, age at onset or incitive studies is the different proportion of women (33%7 and dence of various metabolic disorders. However, we did note that patients with 1 versus more than 1 recurrence were 20%8). The design of our study was prospective but compliance younger at onset of disease. TABLE
Recurrence
~
30
RECURRENCE AFTER FIRST RENAL STONE CONCLUSIONS
Our findings indicate that stones can recur many years after t h e first episode, although our percentage of recurrences was lower t h a n t h at in earlier reports. As even a complete metabolic evaluation including 24-hour urinalysis did not identify patients at risk for recurrence after the first stone episode, it might be worthwhile to evaluate secondary disorders, such as primary hyperparathyroidism and cystinuria. Male patients require close followup as first stone occurred when they were young. The new minimally invasive treatments did not appear to have any appreciable impact on recurrence since in 7 years of followup we found no differences in the incidence of recurrent stones among patients treated in the 5 years before shock wave lithotripsy and those treated after we began to use the less invasive techniques. REFERENCES
1. Williams, R. E.: Long-term survey of 538 patients with upper urinary tract stone. Brit. J. Urol., 3 5 416, 1963. 2. Marshall, V., White, R. H., De Saintonge, M. C., Tresidder, G. C. and Blandy, J. P.: The natural history of renal and ureteric calculi. Brit. J . Urol., 47: 117, 1975. 3. Ljunghall, S. and Hedstrand, H.: Epidemiology of renal stones in a middle-aged male population. Acta Med. Scand., 97: 439, 1975.
4. Ljunghall, S.: Incidence and natural history of renal stone disease and its relationship to calcium metabolism. Eur. Urol., 4: 424, 1978. 5. Sutherland. J . W., Parks, J. H. and Coe, F. L.: Recurrence after a single renal stone in a community practice. Miner. Electrolyte Metab., 11: 267, 1985. 6. Johnson, C. M., Wilson, D. M., OFallon, W. M., Malek, R. S. and Kurland, L. T.: Renal stone epidemiology: a 25-year study in Rochester, Minnesota. Kidney Int., 1 6 624, 1979. 7. Ahlstrand, C. and Tiselius, H. G.: Recurrences during a 10-year follow-up after first renal stone episode. Urol. Res., 18: 397, 1990. 8. Ljunghall. S., Danielson, B. G.: A prospective study of renal stone recurrences. Brit. J. Urol., 56: 122, 1984. 9. Hellstrom, J.: Aetiological and therapeutic experiences concerning kidney and ureteric stones. Brit. J. Urol., 21: 9, 1949. 10. Sutherland, J. W.: Recurrence following operative treatment of upper urinary tract stone. J. Urol., 127: 472, 1982. 11. Scott, R.: Prevalence of calcified upper urinary tract stone disease in a random population survey. Report of a combined study of general practitioners and hospital staff. Brit. J. Urol., 5 9 111, 1987. 12. Pak, C. Y. C . : Should patients with single renal stone occurrence undergo diagnostic evaluation? J. Urol., 127: 855, 1982. 13. Sun, B. Y.-C., Lee, Y.-H., Jiaan, B.-P., Chen, K.-K., Chang, L. S. and Chen, K.-T.: Recurrence rate and risk factors for urinary calculi after extracorporeal shock wave lithotripsy. J. Urol., 156.903, 1996.