Calcium Oxalate Crystalluria: Crystal Size in Urine

0022-5347 /80/1233-0324$02.00/0 Vol. 123, March Printed in U.S.A.

THE JOURNAL OF UROLOGY

Copyright© 1980 by The Williams & Wilkins Co.

CALCIUM OXALATE CRYSTALLURIA: CRYSTAL SIZE IN URINE JAMES S. ELLIOT AND ISRAEL N. RABINOWITZ From the Urology Section, Veterans Administration Medical Center, Palo Alto and the Division of Urology, Stanford University School of Medicine, Stanford, California

ABSTRACT

Studies of calcium oxalate crystals in urine suggest similarities to crystal growth in calcium oxalate renal calculi. Previous reports indicate that urinary crystals in patients in whom stones form are larger than those in normal subjects. We report herein a study on crystal size by structure and habit (shape) based on direct microscopic measurement of crystals in the urine of 27 normal subjects and in 6 of 22 patients in whom stones form. The mean size of all crystals in normal subjects is 12.0 plus or minus 7.8 µm. Calcium oxalate monohydrate crystals are significantly smaller than calcium oxalate dihydrate (p less than 0.01). In 22 patients with stones there was no correlation between crystalluria and severity or duration of disease. The mean crystal size in 6 patients did not support the conclusion that patients in whom stones form excrete larger crystals than normal subjects. We herein report on calcium oxalate crystals in normal urine, or at least urine from patients without stones, and some preliminary observations on crystalluria in patients with known calcium oxalate stone disease. The term normal is used instead of the more cumbersome non-stone patient, although recent estimates have shown that stone disease may affect about 12 per cent of the population 1 and there is no method available to identify a potential patient in whom stones will form. The study was designed primarily to provide data regarding crystal size by habit and structure in urine examined immediately after collection at 37C and again after 4 hours of incubation at 37C, room temperature and in the refrigerator. Crystal size data available in the literature are derived from only a few normal subjects and patients with stones, and refer to all crystals without regard to habit or structure. We have shown that calcium oxalate crystals occur in urine in the dihydrate and the monohydrate habits. 2- 4 The crystal habits (shape) of the dihydrate habit are bipyramids, dodecahedra and interpenetrant twins. The monohydrate habit occurs as biconcave ovals, dumbbells and (rarely) rosettes or forms intermediate in appearance between ovals and dumbbells. The same crystal structures occur in stones with the monohydrate habit, the most frequently occurring and the most probable initiating crystal. 5- 7 The dihydrate habit usually is a surface or secondary deposit on the monohydrate habit. In the central part of stones the monohydrate/dihydrate occurrence ratio is 7:1, 7 whereas in urine the frequency of monohydrate to dihydrate is about 1:6. Crystal habit and growth patterns are similar in stone and urine crystals. In stones the monohydrate habit occurs as thin, needle-like crystals, tending to orient toward a central point and often resulting in a spheroid. This pattern also is seen in urinary crystals described as dumbbells or rosettes. The dihydrate habit occurs in stones as bipyramids, dodecahedra and interpenetrant twins and, except for size and the physical connection between crystals, appear exactly as those seen in the urine. The similarities of crystal habit and growth in urine and stones suggest that further studies on calcium oxalate crystals in urine might be of value in improving our understanding of calcium oxalate calculus formation and growth. In a pioneering study Robertson counted and sized urinary crystals with a Coulter counter, with the urine collected and maintained at 37C. 8 Using this method Robertson and associates found that crystals in normal urine were consistently 3 to 4 µm. in size, whereas patients in whom stones form pass octahedral crystals 10 to 12 µm. in size. 9 Later, Robertson and

Peacock compared 6 patients with calcium oxalate stones to 6 normal subjects and confirmed the difference in size distribution. 10 In 1976 Robertson and associates incorporated crystal size, growth inhibition in metastable salt solutions and the calculated degree of urine supersaturation into a saturationinhibition index as a predictor of stone formation. 11 In all of these innovative studies no distinctions were made between the dihydrate or the monohydrate habits or their various habits. Direct microscopic measurement of the size of individual habits has not been done in single specimens of either normal urine or urine from patients with stones. Although it is known that bipyramidal dihydrate is the most frequently occurring habit the frequency distribution of all habits has not been studied in individual specimens. Although our previous studies have revealed that calcium oxalate crystals occur in about 4.2 per cent of clinical laboratory urine specimens the frequency of crystalluria in stone patients has not been established. In the studies by Robertson and associates the urine was collected at 37C. 8 - 11 However, it is not known whether collection and examination at room temperature are significantly different from collection at 37C with regard to the presence or absence of crystals or whether elapsed time between voiding and examination results in crystal formation, complete dissolution or other changes. This information is needed not only to evaluate reports of the clinical laboratory with regard to the presence or absence of crystals but also to provide data regarding crystal size and habit to compare with experiments in simulated urine, such as reported by Miller and associates. 12 METHODS

Freshly voided urine specimens were collected by medical aides from unselected ambulatory outpatients with no known urological disease. Collections were discontinued after 33 specimens without crystals and 27 specimens with crystals had been acquired, with each specimen collected from a single patient. Specimens were collected from patients with stones in a different fashion. As a stone patient reported to the outpatient clinic for a previously scheduled routine followup visit he was asked to provide at least 3 specimens at intervals of several days. All specimens were collected at 37C in an insulated container. The patients were all men and the age distribution was comparable for normal subjects and patients. A 10 ml. aliquot was withdrawn for immediate initial examination using plain and polarized light microscopy. The remaining specimen was divided into thirds: a third was placed in the incubator at 37C, a third was left at room temperature and a

Accepted for publication June 18, 1979. Supported by the Veterans Administration.

324

325

CALCIUM OXALATE CRYSTALLURIA

third was placed in a refrigerator at 6C. After swirling to ensure complete mixing 10 ml. aliquots were withdrawn from each third at 4 hours. All 10 ml. aliquots were processed as follows: after centrifuging at 2,000 revolutions per minute and removing the supernatant the remaining 0.1 to 0.3 ml. packed crystals and urine were mixed by repeated aspiration in a Pasteur pipette and drop amounts were placed on a slide and covered with a cover slip. Relative habit frequency was determined by differential counts (as in differential white blood counts) and habit size was determined by measuring 20 crystals. Crystal size was measured with a calibrated eyepiece micrometer and recorded as the maximum dimension. No attempt was made to count total crystals per unit volume or to size crystals other than calcium oxalate. In any case, there is no available procedure for counting the extremely small, amorphous uric acid crystals or the submicroscopic particles of apatite. The lower limit of size for measurement and identification using 1,000X magnification was approximately 0.8 µm. In practice crystal sizes were rounded off to the nearest µm. Direct microscopic measurement of crystal size is accepted practice 13' 14 and avoids the error inherent in other procedures when a crystal with unequal dimensions passes through a sieve or aperture and only the shortest may be measured. It also makes possible the sizing of individual habits, which cannot be done by the Coulter counter. RESULTS

1. Crystal combinations in initial specimens of normal urine Occurrence

Bipyramids only Bipyramids and twins Bipyramids, twins, dodecahedra Bipyramids, dodecahedra Dodecahedra, twins Ovals, dumbbells Ovals, dumbbells, bipyramids, twins Ovals, dumbbells, bipyramids, dodecahedra Ovals, bipyramids, twins

1 12 3

1 1 1 1

1 6

27 Totals: Dihydrate present Dihydrate only Monohydrate and dihydrate Monohydrate only

26 18 9

1

Key: Dihydrate: Bipyramids, twins, dodecahedra Monohydrate: Ovals, dumbbells TABLE 2.

Relative frequency of crystal habits in 27 specimens Occurrence (No.)

Bipyramids Twins Multiple interpenetrant twins Dodecahedra Ovals Dumbbells Intermediate * Mean

± standard deviation.

% Total When Present*

25 21 17

66 10

6

43 37 6 24

9

5 1

11

28 7 8 16 32 2

3. Initial crystal size (µm.) in urine of 27 normal patients No. Mean ± Standard Deviation

Dihydrate: Bipyramids Twins Multiple interpenetrant twins Dodecahedra All dihydrate Monohydrate: Ovals Dumbbells Intermediate All monohydrate All crystals

328 40 41

11.5 ± 7.6 12.6 ± 6.1 16.9 ± 10.8

52 461

14.6 ± 8.1 12.4 ± 8.1

68

9.1 7.2 15.4 9.3 12.0

6

5 79 540

5.9 5.6 4.2 6.0 7.8

180

16

14

I"'

~

(.)

0

.8E :J

Normal urine. The combinations and relative frequencies of crystal habits found in 27 freshly voided normal urine specimens are shown in tables 1 and 2. Crystal sizes by structure and habit are given in table 3. The smallest crystal identified was a single 1 µm. monohydrate oval. With this exception, all monohydrate habits ranged from 2 to 34 µm. and the dihydrate habits ranged from 2 to 104 µm. The mean size of all dihydrate habits was significantly larger (p <0.01) than the mean size of all monohydrate habits (see figure). Mean ± standard deviation of crystal size for all crystals in a single specimen ranged from 3.9 ± 2.2 to 24.6 ± 5.9 µm. The median across all specimen means was 11.1 ± 4.0 µm. When all crystals are taken together the proportion of monTABLE

TABLE

z

100

80

6

D

4

Oihydrate

~ Monohydrate

2-6

7-11

12-16

17-21

22-26

27-31

32-36

37-41

>41

JJm

Crystal size distribution in normal urine

ohydrate to dihydrate is 1:5.8. However, in specimens in which the monohydrate proportion is higher the mean crystal size may be affected by the presence of the smaller monohydrate habits. For example, in 1 specimen in which only monohydrate was present the mean crystal size was 8.8 ± 1.5 µm. After incubation for 4 hours at 37, 20 and 6C no crystals formed in the 33 urine samples in which no crystals were present initially. Of the 33 specimens incubated for 4 hours 18 also were incubated for 24 hours. A few crystals appeared in 2 specimens incubated at 20 and 6C. Of the 27 specimens with crystals initially none showed complete dissolution with time. From this fact we conclude that common clinical laboratory procedures are adequate to detect crystalluria. Of the 27 specimens with initial crystalluria 12 showed no change in mean crystal size at any temperature after 4 hours. Significant (p <0.01) increases in mean size occurred in 7 specimens and decreases occurred in 8 (table 4). The greatest percentage size increases occurred at all 3 temperatures in the 3 specimens with the smallest initial size. Decreases occurred only in those specimens with an initial mean size of ~10. 7 µm. We interpret a decrease in mean size to result from the formation of new crystals with or without complete dissolution of larger crystals. Decrease in mean size owing to partial dissolution of crystals is unlikely, since dissolving crystals in urine are quite characteristic with ragged edges and irregular dissolution of crystal faces. Partial dissolution occurred in 1 specimen (No. 19) at 6C coincident with an increased mean size. There was no evident correlation between mean size increases or decreases with crystal habit. Variations in relative propor-

f

f

f

326

ELLIOT AND RABINOWITZ

tions of dihydrate and monohydrate appeared to be characteristic of the individual specimen with no consistent changes resulting from incubation at any temperature. Urine from patients with stones. We obtained 56 specimens from 22 patients with a history of known calcium oxalate stones. The number of specimens obtained from each patient ranged from 1 to 4. The incidence of crystalluria compared to severity and duration of disease is given in table 5. Of the 22 patients only 6 were found to have crystals in the urine and these on 1 occasion only. In 2 of the 6 patients the initial specimen contained ~5 crystals in multiple microscopic fields. In view of the sample size data regarding crystal size and habit distribution 4. Effect of incubation on crystal size in normal urine (mean

TABLE

± standard error) Size After 4 Hrs. (µm.) Pt.

Initial Size 37C

1 2 3 4 6 12 14 15 16

3.9 ± 4.3 ± 4.4 ± 5.7 ± 8.3 ± 10.7 ± 11.1 ± 12.1 ± 12.5 ± 12.6 ± 13.1 ± 13.2 ± 19.4 ± 20.0 ± 24.5 ±

18

19

20 24 25 27

20C

20.8* ± 2.2 1L5* ± LO 8.1* ± 0.6 10.4* ± L7 11.5* ± 0.9 5.7* ± 0.9 20.8* ± 1.1 13.8 ± 1.4 13.4 ±2.0 7.5t ± L3 12.6 ± 0.7 6.lt ± 0.7 14.6 ± L5 14.3t ± 0.9 2LO ± L5

0.5 0.2 0.6 0.6 0.7 1.1 0.9 L5 0.9 0.7 0.3 0.4 L7 L9 L3

12.9* 12.8* 7.9* 2.3 11.3* 5.1* 2L5* 7.8t 8.7t 12.3 12.7 14.7 8.5t 15.7 6.2t

6C

± 0.6 ± 0.8 ± 0.6 ±0.3 ± 0.8 ± 0.6 ± L4 ± 0.7 ± 0.4 ±0.7 ±0.8 ±0.6 ± L2 ±0.6 ± L8

14.0* 9.0* 8.3* 3.9 10.3 7.5 18.5* 14.1 10.2 10.4 15.9* 3.6t 17.6 8.9t 9.lt

± 0.9 ± 0.5 ± 0.5 ±0.4 ±0.9 ± L2 ± 0.9 ± LO ± LO ± LO ± 0.5 ± 0.3 ± 1.4 ± L6 ± 0.7

* Increase ii} size, p <0.0L t Decrease in size, p <0.0L TABLE

Age at Pt.-Onset (yrs.)

5. Incidence of crystals in patients with stone Duration of Disease (yrs.)

HF-25 A0-32 LB-30 RM-25 RS-28 OM-31 NM-33 RC-52 CS-35 EE-43 DD-30 AS-68 JS-78 GG-27 JH-27 JL-47 GW-57 EG-45 LP-30 IM-48 DS-50 JT-32

Specimen No.* No. Stones

16 20

40 30 15 15 13 9 7 5 4 4 2 1 1 1 1 1 1 1 1 1 1 1

19

10 21 16

40 16 20 3 10 2 3 1 2 2 2 1 6

1 6 3

1

2

3

4

0 0 0 0 0 0 0

0 0 0

+

0

+ 0 0 0 0 0 0

+ 0 0 0 0

+ +

0 0 0 0 0 0 0 0

REFERENCES 0 0 0

0

0

+

0 0 0 0

0 0 0 0

0 0

0 0

0 0

0

6. Initial crystal size (µm.) in urine of 6 patients with stone

Dihydrate: Bipyramids Twins Multiple interpenetrant twins Dodecahedra All dihydrate Monohydrate: Ovals Dumbbells Rosettes All monohydrate All crystals

No.

Mean ± Standard Deviation

30 8 3

12.6 ± 6.8 10.1 ± 2.1 20.3 ± 3.2

7 48

10.1 ± 2.5 12.3 ± 6.0

28 6 4 38 86

6.1 7.0 7.7 6.4 9.7

DISCUSSION

Although the incidence of crystalluria in the patients with stones studied did not correlate with the severity of disease crystals were found in 6 of 56 specimens (11 per cent), in 6 of 22 urine samples from patients with stones (27 per cent) and if only the first specimens were counted in 18 per cent. These percentages all are considerably higher than our estimate of 4.2 per cent2 of crystalluria in random specimens. Our finding that crystals in patients with stones are smaller than crystals in normal subjects is at variance with reports that the urine of patients with stones is characterized by larger than normal crystals. Although our data are based on only 6 patients data in the literature are based on 6, 9 • 10 811 or 2 patients. 12 When the range of mean crystal sizes found in our normal urine samples are considered the differences may be owing to inadequate sample sizes or patient selection. Additional studies of urine from patients with stones are needed to establish whether crystal size and habit distribution actually are different from crystals in normal urine. Statistical assistance was provided by Prof. B. W. Brown, Jr., Biostatistics, Stanford University School of Medicine.

0

* 0 denotes crystal-free urine and + denotes urine with crystals. TABLE

must be regarded as preliminary (table 6). When bipyramids alone or all dihydrate habits are compared there is no significant size difference between normal and stone patient crystals. When all monohydrate habits are compared the stone patient crystals are smaller (6.4 versus 9.4 µm., p <0.01). Similarly, when all crystal habits are compared the stone patient crystals are smaller than the crystals in normal urine (9.8 versus 12.0 µm., p <0.01). The largest bipyramid observed was 35 µm. and the largest monohydrate habit was a 21 µm. oval. After incubation of the 50 specimens that contained no crystals initially 1 specimen had 10 crystals in multiple fields at 37C and 1 specimen had multiple crystals at 20 and 6C. Of the 4 specimens containing multiple crystals initially the mean size ranged from 6.4 ± 2.0 to 14.8 ± 5.9 µm. When incubated mean size decreased significantly (p <0.01) at 37C in 1 specimen. Otherwise, there were no significant changes in mean size or habit distribution.

L8 L5 2.0 L8 5.5

1. Sierakowski, R., Finlayson, B., Landes, R. R., Finlayson, C. D. and Sierakowski, N.: The frequency of urolithiasis in hospital discharge diagnoses in the United States. Invest. Urol., 15: 438, 1973. 2. Elliot, J. S., Rabinowitz, I. N. and Silvert, M.: Calcium oxalate crystalluria. J. Urol., 116: 773, 1976. 3. Elliot, J. S. and Rabinowitz, I. N.: Crystal habit, structure and incidence in the urine of a hospital population. In: Urolithiasis Research. Edited by H. Fleisch, W. G. Robertson, L. H. Smith and W. Vahlensieck. New York: Plenum Press, 1976. 4. Elliot, J. S. and Rabinowitz, I. N.: The recognition of urinary crystals. Urol. Digest, p. 13, December 1977. 5. Prien, E. L. and Frondel, C.: Studies in urolithiasis. I. The composition of urinary calculi. J. Urol., 57: 949, 1947. 6. Prien, E. L.: Crystallographic analysis of urinary calculi: a 23-year survey study. J. Urol., 89: 917, 1963. 7. Elliot, J. S.: Structure and composition of urinary calculi. J. Urol., 109: 82, 1973. 8. Robertson, W. G.: A method for measuring calcium crystalluria. Clin. Chim. Acta, 26: 105, 1969. 9. Robertson, W. G., Peacock, M. and Nordin, B. E. C.: Calcium crystalluria in recurrent renal-stone formers. Lancet, 2: 21, 1969. 10. Robertson, W. G. and Peacock, M.: Calcium oxalate crystalluria and inhibitors of crystallization in recurrent renal stone-formers. Clin. Sci., 43: 499, 1972. 11. Robertson, W. G., Peacock, M., Marshall, R. W., Marshall, D. H. and Nordin, B. E. C.: Saturation-inhibition index as a measure of the risk of calcium oxalate stone formation in the urinary tract. New Engl. J. Med., 294: 249, 1976. 12. Miller, J. D., Randolph, A. D. and Drach, G. W.: Observations upon calcium oxalate crystallization kinetics in simulated urine. J.

CALCIUM OXALATE CRYSTALLURIA

Urol., 117: 342, 1977. 13. Mullin, J. W.: Crystallization. Cleveland, Ohio: CRC Press, 1972. 14. Hartman, P. E.: Crystal Growth. New York: American Elsevier Publishing Co., Inc., 1973.

EDITORIAL COMMENT These authors elaborate on the crystal habits of calcium oxalate in normal urine. However, the observations that there is no difference in calcium oxalate dihydrate crystal size between normal subjects and those in whom stones form and that patients who have stones excrete smaller monohydrate crystals than normal subjects are at variance with the previously published studies of Dent and Sutor, 1 and Robertson and Peacock (reference 10 in article). It is important to realize (as

327

mentioned in this study) that only 6 of the 56 urine samples from patients in whom stones form demonstrated crystalluria and that 4 of these were in patients who had had a single calculus. Therefore, these patients differ from those of Robertson in the severity of the disease. The finding that patients in whom stones form excrete smaller crystals, thus, needs substantiation by a larger series. Mani Menon Brady Urological Institute The Johns Hopkins Hospital Baltimore, Maryland 1. Dent, C. E. and Sutor, D. J.: Presence or absence of inhibitor of

calcium oxalate crystal growth in urine of normals and stone formers. Lancet, 2: 775, 1971.