- Email: [email protected]
Caliper and ultrasonographic measurements of bovine testicles and a mathematical formula for determining testicular volume and weight in vivo
ELSEVIER
CALIPER AND ULTRASONOGRAPHIC MEASUREMENTS OF BOVINE TESTICLES AND A MATHEMATICAL FORMULA FOR DETERMINING TESTICULAR VOLUME AND WEIGHT IN VIVO
T. L. Bailey, l R. S. Hudson,2 T. A. Powe, z M. G. Riddell, 2 D. F. Wolfe 2 and R. L. Carson 2 1Virginia-Maryland Regional College of Veterinary Medicine Virginia Tech University, Blacksburg, Virginia 24061 2College of Veterinary Medicine Auburn University, Auburn, AL 36849 Received for publication: December 27, 1995 Accepted: October 22, 1997 ABSTRACT This study quantified the relationship between calibrated caliper and ultrasonographic derived measurements of bovine testicles in vivo with actual testicular length, width, volume and weight. The prelate spheroid formula was tested to accurately predict testicular volume and a modification to predict weight. Ten bulls were employed to derive caliper and ultrasound testicle (n = 20) length and width measurements in vivo. Caliper length measurements were more reliable than ultrasound derived lengths, with correlations of r2= 0.8023; P < 0.05 and r2= 0.5111; P < .05, respectively. Width for both the calipers and ultrasound measurements when compared to actual width measurements were r2= 0.7313; P < 0.05 and r2= 0.8310; P < 0.05, respectively. The prelate spheroid formula is reliable in determining testicle (n = 116) volume (r2= 0.8928; P < 0.05). Testicular volume and weight are highly correlated (r2= 0.9776; P < 0.05); therefore, a modification of the prelate spheroid formula was used to predict weight (r2= 0.9084; P < 0.05) against the actual weight. Caliper-derived length and width measurements used in the prediction of volume and weight had correlation coefficients against actual volume and weight of r2= 0.5497; P < 0.05 and r2= 0.6340; P < 0.05, respectively. Ultrasound in vivo measurements for prediction of testicular volume and testicular weight had a correlation of r2= 0.3276; P < 0.05 and r2= 0.6249; P <0.05, respectively. A testicular (n = 116) length to width ratio of 1.8:1 (SEM = 0.01) was determined for both slaughterhouse and castrated animals. Caliper measurements are reliable, inexpensive and much simpler to obtain than ultrasound determinations for in vivo testicle length, width, volume and weight. The two-dimensional measurement of length and width would be a more accurate predictor of testicle volume and weight than the onedimensional measurement of scrotal circumference (SC), especially in bulls with variation in testicular shape. © 1998 by ElsevierScience Inc.
Key words: bovine, testicle, volume, weight Theriogenology 49:581-594, 1998 © 1998 by Elsevier Science Inc.
0093-691X/98/$19.00 PII S0093-691X(98)00009-0
Theriogenology
582 INTRODUCTION
When a bull is selected for breeding a number of factors are evaluated: breed, conformation, libido and mating ability. In addition, the internal accessory sex organs and external genitalia are examined, and semen quality is evaluated (14). Although it is difficult to predict the reproductive capability of a bull, planned systematic evaluation could eliminate selection by "trial and error." Fertility is a primary economic consideration in the beef cattle industry. The importance of fertility, growth rate and carcass quality has been correlated at the ratio of 10:2:1 (5). Thus fertility is 5 times more important economically than the growth rate and 10 times more important than carcass quality (5). Scrotal and testicular measurements, which are simple and inexpensive to obtain, have been used to predict sperm production and semen quality (6). However, because SC is an indirect measurement of testicular mass which discounts variation in scrotal wall thickness and individual testis shape, the length and width measurements could provide a useful tool for predicting bull fertility. Variations in testicular shape or differences in the testicular length to width ratio are commonly observed (10,11,1). The conventional measurement of SC may prove to be inadequate in determining actual weight and volume in these bulls. The establishment of breeding soundness data using SC was primarily based on bulls with ovoid-shaped testes (4). Bulls with more rounded testes that are shorter from pole to pole, may have an unduly high SC score, whereas bulls with long slender testes may have a lower score (1). In all but one (1) previous trial, a relationship was established between scrotal circumference, testicular weight and volume, and daily sperm output, with all authors discounting possible variation in testicular shape. SC and testicular weight are highly correlated (7), but these determinations in the live bull may only be subjective at best. Based on previous caliper and ultrasonography measurements of the boar and bull testis (2,3), a noninvasive method of direct measurement could be achieved. Testicular volume has also been previously calculated with the formula for a cylinder, where V = (FI)(r2)(H) (7,8). A prolate spheroid (12) more closely resembles the normal shape of the cylindrical, round tapered ends of the bovine testicle. A simple mathematical formula using the volume for a prolate spheroid (12,17), where V = 4/3(FI)(L/2)(W/2) 2 and L = length and W = width of a single testicle, was used in this study for calculating testicular volume (12,17). A modified derivation of the prolate spheroid formula, where W = 0.5533(L)(W) 2, was used to predict testicular weight. While several studies have reported that SC is dependent on age, breed, and body condition, they have also indicated that these parameters may influence semen quality more than SC (7,9,15). Other investigators have reported a correlation between SC and seminiferous epithelium (16). Only one study (1), has reported data on measurements of actual testicular width and length and the possible relationship between testicular shape and sperm output. In 1957 Willet and Ohms (18) concluded that measurement of the length of each testicle contributed no additional information to that obtained by the SC measurement. Later, the premise of the 1957 study (18) became the basis of subsequent work in the area of SC and its relationship to
Theriogenology
583
breeding soundness evaluations. Cartee et al. (2) also reported no correlation between testicular length and any other physical measurement (2). The objectives of our present study were 1) to test the formula for a prolate spheroid for determining testicular volume and its modification for testicular weight; 2) to compare the reliability of caliper and ultrasonographic measurements of testicular length and width in vivo; 3) to determine the validity of using caliper and ultrasound measurements for testicular length and width in vivo, applying these measurements in the calculation of testicular volume and weight, and verification of these results with post-excision testicular measurements; and 4) to determine the individual testicular length to width ratio for the normal ovoid-shaped testicular mass. MATERIALS AND METHODS Individual testicles were collected from a local abattoir. The origin and reproductive histories of the 48 bulls from which these testicles were obtained were unknown. The bulls were of different breeds and crossbreeds, and selection was based primarily on their mature phenotype (weight range: 450 to more than 900 kg). Although the bulls varied greatly in size, all were of mature breeding age, as indicated by incisor eruption and wear. Approximately 15 min after each bull had been euthanized, both testicles were collected and the epididymis and pampiniform plexus removed, leaving only the testis proper. Length of the testis was then measured in centimeters with a pair of calibrated calipers from the proximal to distal pole. Width was also measured with the calipers after suspension of the testicle in a waterbath with a pair of towel clamps to mimic the testicle in vivo, and to eliminate distortion of testicle shape as a result of resting on a table. Measurements were made to the nearest 0.1 cm. The volume of each testicle was obtained by the water displacement method, as previously described (7,8). Each testicle was placed in a 2-L graduated cylinder with a known volume of water. The water displaced by each testicle was then read to the nearest 5 ml to determine individual testicular volume. The weight of each testicle was measured to the nearest 0.5 g using a triple beam balance. After recording the actual testicular length, width, volume and weight of the slaughterhouse specimens, formulas for the prediction of testicular volume and weight were tested ( 12,13,17). The predicted volume of each testis of a pair was calculated by the formula for a prolate spheroid (12,17): V = 4/3(H)(L/2)(W/2) 2 , where L equals the length of the testis, and W is the widest diameter of the testis. This formula reduces to: V = 0.5236(L)(W) 2. The product of the calculations from the formula for a prolate spheroid was then tested against the actual volume for verification of the predicted volume of each testis by the least-squares method of fitting a regression line to the data (13). For the regression lines, the equation is in the standard line form: y = a + bx, where b is the slope and a is the y-intercept. To correlate testis measurements for volume and weight, an independent regression analysis for both actual volume and weight was calculated for each testis collected (13). This presented a high correlation between volume and weight of each testicle, and a revised formula was calculated to predict testicular weight: weight = 0.5533(L)(W) 2, where L = length of the testicle and W = width of the testicle. This predicted weight was then compared to the actual testicular weight by the leastsquared method of fitting a regression line to the data (13).
584
Theriogenology
Ten clinically normal bulls with a mean age of 3 yr (range: 1.5 to 4.5 yr) were purchased from a local dealer. Although there were variations in age, the bulls were of sufficient breeding age, as indicated by incisor eruption and SC. All the bulls were maintained in a paddock and were fed free choice coastal bermuda grass hay and 4.5 kg of a 20% protein grain concentrate, and given water ad libitum. Prior to being used in the study, the bulls were subjected to a breeding soundness evaluation (BSE) and determined to be satisfactory potential breeders (14). Measurements were made in vivo with calibrated calipers to obtain individual testicle length and width. Length measurements were taken of the testis proper and excluded the corpus and caudal epididymis. Width was measured with the calipers at the point of maximum scrotal distention to the nearest 0.1 cm, in an anterior to posterior fashion rather than medial to lateral in order to exclude the body of the epididymis and the vas deferens. Gentle downward pressure was applied at the neck of the scrotum for maximum distention. A portable ultrasound unit with a 5 megahertz linear-array transducer was used for ultrasonographic measurements. Each testicle was independently measured for both length and width with the epididymis excluded. The thin scrotal skin and tunics, the constraint of the 12cm ultrasound screen in relation to testicular length, and the echo pattern produced by the testicle made it difficult to measure the entire testicular length. A 4-cm thick sonolucent pad or "standoff" pad was placed between the scrotal skin and the ultrasound probe in an attempt to view the entire testicle and improve image resolution. However, little advantage was gained in viewing the length of the testicle from proximal to distal pole. A "gridline" was fashioned over the sonolucent pad and placed between this pad and the scrotal skin. The testicle was then measured from the proximal pole to a gridline at the approximate midline of the testicle and then from this midline to the distal end, with a sum of the 2 values to attain a length measurement. All measurements were read to the nearest 0.1 cm utilizing the computer-assisted scale on the ultrasound screen. All of the aforementioned phases of the caliper and ultrasound measurements in vivo were repeated 3 times at 3-wk intervals. To reduce variability in technique between persons obtaining the measurements, one person independently performed all caliper evaluations and another all ultrasound-assisted measurements for the study. Following completion of this portion of the trial, the testicles were surgically removed to obtain the actual measurements of the length and width by the method previously described in objective one of the study. All bulls were restrained in a squeeze chute and the scrotum anesthetized with local infiltration of 2% lidocaine hydrochloride. A vertical incision across the ventral aspect of the scrotum was made with a Newberry knife. The common vaginal tunic was incised, ligation of the vasculature was done with No. 2 chromic gut, and the testicle was removed with the accompanying adnexa. The epididymis and pampiniform plexus were removed before the measurements of testicular length and width were taken, as previously described. In addition to length and width, testicular volume and weight were measured by the water displacement method and a balance beam scale. All measurements were concluded within 10 rain of testicular excision. For testicular length and width from the caliper and ultrasound measurements, a mean value from the 3 independent measurements for each bull was calculated. The correlation coefficients and P values were then derived from these mean values and were compared with
Theriogenology
585
actual testicular length and width measurements after excision. Correlation coefficients were also calculated to test actual testicular volume and weight against the prolate spheroid formula for volume and its alteration for the prediction of weight obtained from measurements of length and width with both the caliper and ultrasound measurements in vivo. (Table 1) Table 1. Prediction of Testicle Volumes and Weights from Measurements Taken In Vivo for Length and Width with Calipers and Ultrasound. (Example demonstrates data from 2 bulls ) Bull Number 14 15 Bull Number
Actual Length __ 11.8 14.0
Actual Width 6.6 7.0
Actual Predicted Testicle Ultrasound Volume Volume 14 260 236.9 15 290 245.5 Predicted Volume = 0.5236(L)(W) "~ Predicted Weight = 0.5533(L)(W) 2
Ultrasound Length 11.4 11.1
Ultrasound Width 6.3 6.5
Caliper _Length 11.7 13.1
Caliper Width 6.9 7.5
Predicted Caliper Volume 291.7 385.8
Actual Testicle Weight 280 315
Predicted Ultrasound Weight 250.3 259.5
Predicted Caliper _W_eight 308.2 407.7 i
A length-to-width ratio was calculated for each individual testicle (n = 116) obtained from the slaughterhouse and from the post-excision trial. A subjective evaluation was made of the testicular shape prior to castration, with all bulls having a normal ovoid-shaped scrotal mass. RESULTS The reliability of the formula for a prolate spheroid to calculate testicular volume was tested and verified with excised testicles (n = 116). A correlation coefficient was calculated from actual testicular volume by the water displacement method for each testicle and the predicted volume from the formula for a prolate spheroid. The predictive value for calculating the volume from the formula of prolate spheroid is accurate and reliable, with a correlation coefficient (Figure 1) of rZ= 0.8928 (P < 0.05). A second set of linear regressions was determined for actual testicular volume and testicular weight (Figure 2). The significant correlation coefficient of r2= 0.9776 (P < 0.05) indicates a high degree of relationship between testicular volume and weight. Since volume and weight of the individual testicles were highly correlated, a calculated mathematical alteration in the formula for the volume of a prolate spheroid was tested for reliability in predicting testicular weight from measurements of both the length and width of individual testicles. A correlation coefficient (Figure 3) of r2= 0.9084 (P < 0.05) was achieved when testing predicted weight from the derived formula and actual testicular weight. The previously reported formula of V = (FI)(r2)(H) for calculating the volume of a cylinder used in the computation of testicular volume was also tested (7,8). A correlation coefficient of r 2 = 0.5364 (P < 0.05) was obtained using testicular length and width of excised testicles with the
Theriogenology
586
formula for cylinder volume. This indicates a more reliable (r2= 0.8928; P < 0.05) predictor using the prolate spheroid formula for the calculation of testicle volume. Caliper and ultrasonographic measurements obtained in vivo for testicular length and width were evaluated and verified. In vivo caliper measurements for length compared with the actual testicular length (Figure 4) revealed a correlation coefficient of r2= 0.8023 (P < 0.05). Y = 1.0961(X) - 7.5094
600.0
r 2 = 0.8928
500.0
_= O > 400.0 .~ 0)
~,
O
n = 116
6_ _ q ~ o
300.0
O ~ o
°°
m
400
500
ty-
200.0 100.0 0.0 0
100
200
300
Actual Testicle Volume (cc's) Figure 1. Linear regression and correlation coefficient for actual testicle volume compared with the predicted testicle volume obtained from the formula for a prolate spheroid.
~o
Y = 1.0725(X) + 4.4083
500
r 2 = 0.9776
400 O
>.
t~
_
~
300
•,~
200
E-~
100
<
n = 116
t~,~r°
j
J
0 0
_
I 100
I
I
I
200 300 400 Actual Testicle Weight (gms)
500
Figure 2. Linear regression and correlation coefficient for actual testicle weight compared with actual testicle volume.
Theriogonology
587 Y =10726(X) - 9.8923
600
r 2 = 0.9084 y:
500
n = 116
400 "U 300
¢ 2oo .~
100 0
,,r
,
0
I
F
I
100 200 300 400 Actual Testicle Weight (gms)
500
Figure 3. Linear regression and correlation coefficient for actual testicle weight compared with the predicted testicle weight obtained from the modified formula for a prolate spheroid.
Y = 0.9976(X) - 0.2817 r 2 = 0.8023
"~ 16 a~ 14 = 12 10 qD 8 '~ 6
¢ e~
n=')0
4 2 I
0
2
4 6 8 10 12 Actual Testicle Length (cms)
14
Figure 4. Linear regression and correlation coefficient for actual testicle length and caliper acquired testicle lengths.
This same instrument for width (Figure 5) demonstrated a slightly lower correlation coefficient of r2= 0.7313 (P < 0.05). Ultrasonographic lengths and widths suggest that width is much more
Theriogenology
588
reliable than length when taken with the ultrasound instrument, length correlation coefficients of r2= 0.5111 (P < 0.05) for the ultrasound images were calculated (Figure 6) and compared with actual testicular lengths. Width correlation coefficients (Figure 7), however, were slightly more reliable, with values of r2= 0.8310 (P < 0.05). This demonstrates the degree of difficulty in obtaining accurate length measurements because of limitations with the ultrasound unit reproducing the entire testicle length on the screen and the curvilinear shape of the testicle.
~ 8.0 Y = 0.7449(X) + 2.0937 ~ 6.0 [ r 2 = 0 " 7 3 1 3 ....
4.o 2.0 0.0
I 0
I
t
2 4 6 A c t u a l Testicle W i d t h (cms)
Figure 5. L i n e a r regression and correlation coefficient for actual testicle width and caliper derived testicle width.
,~o 12.0 -cA 10.0 ©
u
Y = 0.5274(X) + 4.4133 r 2 = 0.5111 n = 20
~
~
o
~
8.0 6.0 4.0 2.0 o.o
I
I
~
I
I
L
I
2
4
6
8
10
12
14
Actual Testicle Length (cms) Figure 6. Linear regression and correlation coefficient for actual testicle length compared to testicle lengths obtained by ultrasound.
Theriogenology
589
Y = 0.6707(X) - 1.9171 *~
8.0
r 2 = 0.8310
60 -~.~
4.0
~:
2.0
0
2 4 6 Actual Testicle Width (cms)
Figure 7. Linear regression and correlation coefficient for actual testicle width c o m p a r e d to ultrasound obtained widths. Caliper and ultrasonography measurements for both length and width from in vivo testicles were used with the formulas for prediction of testicular volume and weight. The data suggests a slightly higher degree of reliability was achieved when measurements were obtained with the calipers. A correlation coefficient (Figure 8) of r2= 0.5497 (P < 0.05) was determined when calculating predicted volume from caliper measurements in vivo.
Y = 0.990
450
400 350
[-., .-.. "o ,~
300
"
250
"-"
( X ) + 38.55
r 2 = 0.5497 n=20
o
200 ~, ~
150
>
100
,,.-,
=
f
50
I
0
0
I
I
100 200 300 Actual Testicle V o l u m e (ccs)
400
Figure 8. Linear regression and correlation coefficient for actual testicle volume versus predicted testicle v o l u m e for length and width measurements obtained by caliper measurements.
Thedogeno~gy
590
A lower correlation coefficient (Figure 9) of ta= 0.3276 (P < 0.05) was derived for predicted testicular volume from measurements when using ultrasonographically obtained lengths and widths of testicles in vivo. The correlation coefficients achieved for predicted testicular weight with the respective instruments, where r2= 0.6340 (P< 0.05) for the calipers (Figure 10) and r2= 0.6249 (P < 0.05) for the ultrasound unit (Figure 11), indicate both instruments have essentially the same degree of accuracy in predicting testicle weight. Y = 0.3915(X) + 120.5
o
-o
300
[-
250
.~ ~~
200
"~ " "
150
.~_
100
•=
I:= O
~
>
r2 = 0.3276 n=20 o
O
o
50
D
0
1 O0
200
300
400
Actual Testicle Volume (ccs) Figure 9. Linear regression and correlation coefficient for actual testicle volume compared to predicted testicle volume for length and width measurements obtained by ultrasound.
Y = 1.275(x) -33.91 r2 = 0 6340 _-_= 20
450
"- ~
400
[...., .~--.-
350 300
•~ ~:
200 150 100 50
z 250
,~
=~
O
~
-
f v v
/-
o 0
50
100 150 2 0 0 2 5 0 3 0 0 3 5 0 4 0 0 Actual Testicle Weight (gm)
Figure 10. Linear regression and correlation coefficient for actual testicle weight compared to predicted testicle weight for length and width measurements obtained caliper length and width measurements.
Thefiogeno~gy
591
Y = 0.550(X) + 85.48
400 t.3 •=r~ 350 300 ,.-. 250
r 2 = 0.6249 n = 20
lOO f
~:
50 o
J
0
50
~
L
I
I
I
100 150 200 250 300 Actual Testicle Weight (gms)
350
Figure 11. Linear regression and correlation coefficient for actual testicle weight compared to predicted testicle weight for length and width measurements obtained by ultrasound. We also calculated from both populations of bulls (n = 116) in this project, a postexcision ratio of testicular length to width at 1.81:1 (SEM = 0.01). These results were obtained from bulls which had, based on subjective evaluation, a normal ovoid-shaped scrotum and testicular mass. DISCUSSION The formula for a prolate spheroid and its modification provides a rapid and highly reliable method for estimating testicular volume and weight from testicular length and width measurements. The advantages of length and width measurements and application to a simple mathematical formula are readily apparent. This formula utilizes real physical measurements of individual testicutar length and width, unlike that of SC which measures only 1 dimension and encompasses both testicles simultaneously. The formula would be useful in following an animal through a long term course of study for an accurate assessment of single testicle volume or weight changes with time. Bulls are often presented with a variation in testicular shape (1,10,11). Frequently seen are the long slender testicles, rounded testicles, and normal ovoid scrotum (1,10,11). The physical measurement of length and width can now be considered in the calculation of total testicular volume and weight, rather than a single measurement of SC, or testicular width, which may not reflect the total volume or weight. Since individual testicles can be measured and the formula applied, variation in testicular shape among individual bulls is discounted as a potential problem in calculating total volume or weight of the 2 testicles. Previous reports of formulas and regression equations for calculating testicular volume from in vivo length and width measurements were given and tested (7,8). The formula of V = (FI)(r2)(H) for computing the volume of a cylinder is referenced and used in the calculation of testicular volume (7,8). The bovine testicle is not synonymous with the shape of a cylinder (12).
592
Theriogenology
However, a prolate spheroid more closely resembles the true shape of the testicle and would therefore be more appropriate for computation of testicular volume (12,17). It is also important to note that volume is a real physical measurement of the testis, subject only to measurement error (17). The data from the regression equation is linear over the entire range of predicted testicular volumes and weights calculated against the actual measurements. This would suggest that testicular measurements are accurate in predicting the volume of testicles at extreme lengths and widths encountered in adult bulls. Regression analysis demonstrated testicular volume and weight are highly correlated. Therefore, the prolate spheroid formula for predicting volume could be easily converted for accurate determination of testicular weight. This was determined from mean weights and volumes, and the calculation of a constant for the weight formula, as 0.5236 is for the volume formula, where V = 0.5236(L)(W) 2. A factor of 0.5533 was found to be valid for the weight formulation, where testicular weight = 0.5533(L)(W) 2 . Analysis showed that measurement of testicular widths and lengths were good predictors of testicular weight; therefore, weight could be determined from these physical dimensions. The data demonstrates the reliability of using caliper and ultrasonographic imaging for measurements of testicular length and width. Both instruments are quite reliable for width measurements; however, calipers are easier to use and demonstrate a higher degree of accuracy for measurement of testicular length. Further work is needed to alleviate the problems encountered determining testicular lengths by ultrasound. This data suggests that caliper measurements for determining the length and width and subsequent calculation of testicular volume and weight would be a reliable method for use in bulls for individual testicles lying outside the normal ovoid parameters. Our justification for this assumption is that 2 physical measurements are taken, rather than a single measurement of the SC or testicular diameter. This would include both the bulls with the more rounded, spheroid-shaped testicles and those with long, slender testicles (1,10,11). Caliper and ultrasonographic measurements for length and width of the testicles had an acceptable correlation in the calculation of testicular weight. However, ultrasound values for the measurement of length and width were not as well correlated with the actual testicular volume. Difficulty was encountered when measuring a curvilinear structure in a 2 dimensional plane and accurately determining length. The computer-assisted scale contained within the ultrasound unit had a constraint of approximately 12 cm; therefore, testicles greater than 12 centimeters in length would be incomplete from pole to pole on the screen. To try and circumvent this problem, a "stand- off" pad was placed between the probe and the scrotum; however, little advantage was gained in viewing the testicle in its entirety. A gridline pad was also applied between the scrotum and the probe to measure from the cranial pole of the testicle to a marked midline on the grid, then from this midline to the caudal pole. This was cumbersome and produced great difficulty in obtaining accurate measurements. We conclude that ultrasound is applicable in diagnostics, but it adds no advantage over that of calipers for measuring the length of the in vivo testicles except when only a width measurement is desired. In the estimation of total weight or volume, both testicles should be measured for length and width to alleviate error in calculation because of possible disparate size of the 2 testicles. However, if the testicles are of equal size, one can be measured and the formula product multiplied by 2 for total testicular volume or weight.
Theriogenology
593
We conclude that the formula for a prolate spheroid, utilizing measurements for the dimensions of length and width, would be more accurate in determining testicular volume and the revised formula for weight of the long, slender or more rounded testicle (1). This formulation relies on 2 measurements for calculation, rather than only the one of width utilized in previous studies for the formulation of predicted volume and weight. We submit a more accurate assessment would be obtained from 2 physical measurements, especially in bulls with a variation in testicular shape. However, more research is needed in bulls with this variation. The ratio of testicular length to width in mature bulls with normal ovoid-shaped testicles is 1.8:1. This data is based on excised testicles and on the derivation of length-to-width ratios. This would be useful in objectively determining the physical shape of the individual bovine testicle. Again, we evaluated only those bulls with a normal ovoid-shaped testicular mass; therefore, more data needs to be collected from bulls with a variation in testicular shape. REFERENCES 1. Bailey TL, Monke DR, Hudson RS, Wolfe DF, Carson RL, and Riddell G. Testicular shape and its relationship to sperm production in mature Holstein bulls. Theriogenology 1996, 46:881-887. 2. Cartee RE, Gray BW, Powe TA, Hudson RS, Whitesides J. Preliminary implications of BMode ultrasonography of the testicles of beef bulls with normal breeding soundness examinations. Theriogenology 1989;31:1149-1157. 3. Cartee RE, Powe TA, Gray BW, Hudson RS, Kuhlers DL. Ultrasonographic examination of the boar testicle. Am J Vet Res 1986;47:2543-2548. 4. Cates WF. Observations on Scrotal Circumference and Its Relationship to Classification of Bulls. Proc Ann Meeting of the Soc for Theriogenology 1975; 3-18. 5. Coulter GH. Puberty and postpubertal development of beef bulls, in Morrow DA (ed), Current Therapy in Theriogenology. Philadelphia: WB Saunders Company, 1986; 142-148. 6. Elmore RG, Bierschwal CJ, Youngquist RS. Scrotal circumference measurements in 764 beef bulls. Theriogenology 1976;6:485-494. 7. Fields MJ, Burns WC, Warnick AC. Age, season and breed effects on testicular volume and semen traits in young beef bulls. J Anim Sci 1979;48:1229-1334. 8. Hahn JR, Foot RH, Seidel GE. Testicular growth and related sperm output in dairy bulls. J Anim Sci 1969;29:41. 9. Lunstra DD, Echternkamp SE. Puberty in beef bulls: acrosome morphology and semen quality in bulls of different breeds. J Anim Sci 1982;55:638-648. 10. Monke D. Examination of the bovine scrotum, testicles, and epididymides. Part I. Comp Cont Ed Pract Vet 1987;9:F252-F258. 11. Monke D. Examination of the bovine scrotum, testicles, and epididymides. Part II Comp Cont Ed Pract Vet 1987;9:F277-F283. 12. Morrill WK, Selby SM, Johnson WG. Surfaces and Curves. In: Modem Analytical Geometry. Scranton, Pa: Haddon Craftsman, Inc., 1972;398-432. 13. SAS Introductory Guide for Personal Computers. Cary NC: SAS Institute Inc, 1985. 14. Society for Theriogenology: Manual for Breeding Soundness Examination of Bulls. 1983;12. 15. Thompson HA, Johnson WH. Scrotal size of yearlings sires and early calving in beef herds: Epidemiological investigation of possible causal pathways. Theriogenology 1995;43:12791287.
594
Theriogenology
16. Veeramachaneni DNR, Ott RS, Heath EH, McEntee K, Bolt D J, Hixon, JE. Pathophysiology of small testes in beef bulls; relationship between scrotal circumference, histopathologic features of testes and epididymis, seminal characteristics and endocrine profiles. Am J Vet Res 1986;47:1988-1999. 17.Watson-Whitmyre M, Stetson MH. A mathematical model for estimating paired testes weight from in situ testicular measurements in three species of hamster. Anat Rec 1985;213:473-476. 18.Willet EL, Ohms JI. Measurements of testicular size and its relationship to production of spermatozoa by bulls. J Dairy Sci 1957;40:1559-1569.
CALIPER AND ULTRASONOGRAPHIC MEASUREMENTS OF BOVINE TESTICLES AND A MATHEMATICAL FORMULA FOR DETERMINING TESTICULAR VOLUME AND WEIGHT IN VIVO
T. L. Bailey, l R. S. Hudson,2 T. A. Powe, z M. G. Riddell, 2 D. F. Wolfe 2 and R. L. Carson 2 1Virginia-Maryland Regional College of Veterinary Medicine Virginia Tech University, Blacksburg, Virginia 24061 2College of Veterinary Medicine Auburn University, Auburn, AL 36849 Received for publication: December 27, 1995 Accepted: October 22, 1997 ABSTRACT This study quantified the relationship between calibrated caliper and ultrasonographic derived measurements of bovine testicles in vivo with actual testicular length, width, volume and weight. The prelate spheroid formula was tested to accurately predict testicular volume and a modification to predict weight. Ten bulls were employed to derive caliper and ultrasound testicle (n = 20) length and width measurements in vivo. Caliper length measurements were more reliable than ultrasound derived lengths, with correlations of r2= 0.8023; P < 0.05 and r2= 0.5111; P < .05, respectively. Width for both the calipers and ultrasound measurements when compared to actual width measurements were r2= 0.7313; P < 0.05 and r2= 0.8310; P < 0.05, respectively. The prelate spheroid formula is reliable in determining testicle (n = 116) volume (r2= 0.8928; P < 0.05). Testicular volume and weight are highly correlated (r2= 0.9776; P < 0.05); therefore, a modification of the prelate spheroid formula was used to predict weight (r2= 0.9084; P < 0.05) against the actual weight. Caliper-derived length and width measurements used in the prediction of volume and weight had correlation coefficients against actual volume and weight of r2= 0.5497; P < 0.05 and r2= 0.6340; P < 0.05, respectively. Ultrasound in vivo measurements for prediction of testicular volume and testicular weight had a correlation of r2= 0.3276; P < 0.05 and r2= 0.6249; P <0.05, respectively. A testicular (n = 116) length to width ratio of 1.8:1 (SEM = 0.01) was determined for both slaughterhouse and castrated animals. Caliper measurements are reliable, inexpensive and much simpler to obtain than ultrasound determinations for in vivo testicle length, width, volume and weight. The two-dimensional measurement of length and width would be a more accurate predictor of testicle volume and weight than the onedimensional measurement of scrotal circumference (SC), especially in bulls with variation in testicular shape. © 1998 by ElsevierScience Inc.
Key words: bovine, testicle, volume, weight Theriogenology 49:581-594, 1998 © 1998 by Elsevier Science Inc.
0093-691X/98/$19.00 PII S0093-691X(98)00009-0
Theriogenology
582 INTRODUCTION
When a bull is selected for breeding a number of factors are evaluated: breed, conformation, libido and mating ability. In addition, the internal accessory sex organs and external genitalia are examined, and semen quality is evaluated (14). Although it is difficult to predict the reproductive capability of a bull, planned systematic evaluation could eliminate selection by "trial and error." Fertility is a primary economic consideration in the beef cattle industry. The importance of fertility, growth rate and carcass quality has been correlated at the ratio of 10:2:1 (5). Thus fertility is 5 times more important economically than the growth rate and 10 times more important than carcass quality (5). Scrotal and testicular measurements, which are simple and inexpensive to obtain, have been used to predict sperm production and semen quality (6). However, because SC is an indirect measurement of testicular mass which discounts variation in scrotal wall thickness and individual testis shape, the length and width measurements could provide a useful tool for predicting bull fertility. Variations in testicular shape or differences in the testicular length to width ratio are commonly observed (10,11,1). The conventional measurement of SC may prove to be inadequate in determining actual weight and volume in these bulls. The establishment of breeding soundness data using SC was primarily based on bulls with ovoid-shaped testes (4). Bulls with more rounded testes that are shorter from pole to pole, may have an unduly high SC score, whereas bulls with long slender testes may have a lower score (1). In all but one (1) previous trial, a relationship was established between scrotal circumference, testicular weight and volume, and daily sperm output, with all authors discounting possible variation in testicular shape. SC and testicular weight are highly correlated (7), but these determinations in the live bull may only be subjective at best. Based on previous caliper and ultrasonography measurements of the boar and bull testis (2,3), a noninvasive method of direct measurement could be achieved. Testicular volume has also been previously calculated with the formula for a cylinder, where V = (FI)(r2)(H) (7,8). A prolate spheroid (12) more closely resembles the normal shape of the cylindrical, round tapered ends of the bovine testicle. A simple mathematical formula using the volume for a prolate spheroid (12,17), where V = 4/3(FI)(L/2)(W/2) 2 and L = length and W = width of a single testicle, was used in this study for calculating testicular volume (12,17). A modified derivation of the prolate spheroid formula, where W = 0.5533(L)(W) 2, was used to predict testicular weight. While several studies have reported that SC is dependent on age, breed, and body condition, they have also indicated that these parameters may influence semen quality more than SC (7,9,15). Other investigators have reported a correlation between SC and seminiferous epithelium (16). Only one study (1), has reported data on measurements of actual testicular width and length and the possible relationship between testicular shape and sperm output. In 1957 Willet and Ohms (18) concluded that measurement of the length of each testicle contributed no additional information to that obtained by the SC measurement. Later, the premise of the 1957 study (18) became the basis of subsequent work in the area of SC and its relationship to
Theriogenology
583
breeding soundness evaluations. Cartee et al. (2) also reported no correlation between testicular length and any other physical measurement (2). The objectives of our present study were 1) to test the formula for a prolate spheroid for determining testicular volume and its modification for testicular weight; 2) to compare the reliability of caliper and ultrasonographic measurements of testicular length and width in vivo; 3) to determine the validity of using caliper and ultrasound measurements for testicular length and width in vivo, applying these measurements in the calculation of testicular volume and weight, and verification of these results with post-excision testicular measurements; and 4) to determine the individual testicular length to width ratio for the normal ovoid-shaped testicular mass. MATERIALS AND METHODS Individual testicles were collected from a local abattoir. The origin and reproductive histories of the 48 bulls from which these testicles were obtained were unknown. The bulls were of different breeds and crossbreeds, and selection was based primarily on their mature phenotype (weight range: 450 to more than 900 kg). Although the bulls varied greatly in size, all were of mature breeding age, as indicated by incisor eruption and wear. Approximately 15 min after each bull had been euthanized, both testicles were collected and the epididymis and pampiniform plexus removed, leaving only the testis proper. Length of the testis was then measured in centimeters with a pair of calibrated calipers from the proximal to distal pole. Width was also measured with the calipers after suspension of the testicle in a waterbath with a pair of towel clamps to mimic the testicle in vivo, and to eliminate distortion of testicle shape as a result of resting on a table. Measurements were made to the nearest 0.1 cm. The volume of each testicle was obtained by the water displacement method, as previously described (7,8). Each testicle was placed in a 2-L graduated cylinder with a known volume of water. The water displaced by each testicle was then read to the nearest 5 ml to determine individual testicular volume. The weight of each testicle was measured to the nearest 0.5 g using a triple beam balance. After recording the actual testicular length, width, volume and weight of the slaughterhouse specimens, formulas for the prediction of testicular volume and weight were tested ( 12,13,17). The predicted volume of each testis of a pair was calculated by the formula for a prolate spheroid (12,17): V = 4/3(H)(L/2)(W/2) 2 , where L equals the length of the testis, and W is the widest diameter of the testis. This formula reduces to: V = 0.5236(L)(W) 2. The product of the calculations from the formula for a prolate spheroid was then tested against the actual volume for verification of the predicted volume of each testis by the least-squares method of fitting a regression line to the data (13). For the regression lines, the equation is in the standard line form: y = a + bx, where b is the slope and a is the y-intercept. To correlate testis measurements for volume and weight, an independent regression analysis for both actual volume and weight was calculated for each testis collected (13). This presented a high correlation between volume and weight of each testicle, and a revised formula was calculated to predict testicular weight: weight = 0.5533(L)(W) 2, where L = length of the testicle and W = width of the testicle. This predicted weight was then compared to the actual testicular weight by the leastsquared method of fitting a regression line to the data (13).
584
Theriogenology
Ten clinically normal bulls with a mean age of 3 yr (range: 1.5 to 4.5 yr) were purchased from a local dealer. Although there were variations in age, the bulls were of sufficient breeding age, as indicated by incisor eruption and SC. All the bulls were maintained in a paddock and were fed free choice coastal bermuda grass hay and 4.5 kg of a 20% protein grain concentrate, and given water ad libitum. Prior to being used in the study, the bulls were subjected to a breeding soundness evaluation (BSE) and determined to be satisfactory potential breeders (14). Measurements were made in vivo with calibrated calipers to obtain individual testicle length and width. Length measurements were taken of the testis proper and excluded the corpus and caudal epididymis. Width was measured with the calipers at the point of maximum scrotal distention to the nearest 0.1 cm, in an anterior to posterior fashion rather than medial to lateral in order to exclude the body of the epididymis and the vas deferens. Gentle downward pressure was applied at the neck of the scrotum for maximum distention. A portable ultrasound unit with a 5 megahertz linear-array transducer was used for ultrasonographic measurements. Each testicle was independently measured for both length and width with the epididymis excluded. The thin scrotal skin and tunics, the constraint of the 12cm ultrasound screen in relation to testicular length, and the echo pattern produced by the testicle made it difficult to measure the entire testicular length. A 4-cm thick sonolucent pad or "standoff" pad was placed between the scrotal skin and the ultrasound probe in an attempt to view the entire testicle and improve image resolution. However, little advantage was gained in viewing the length of the testicle from proximal to distal pole. A "gridline" was fashioned over the sonolucent pad and placed between this pad and the scrotal skin. The testicle was then measured from the proximal pole to a gridline at the approximate midline of the testicle and then from this midline to the distal end, with a sum of the 2 values to attain a length measurement. All measurements were read to the nearest 0.1 cm utilizing the computer-assisted scale on the ultrasound screen. All of the aforementioned phases of the caliper and ultrasound measurements in vivo were repeated 3 times at 3-wk intervals. To reduce variability in technique between persons obtaining the measurements, one person independently performed all caliper evaluations and another all ultrasound-assisted measurements for the study. Following completion of this portion of the trial, the testicles were surgically removed to obtain the actual measurements of the length and width by the method previously described in objective one of the study. All bulls were restrained in a squeeze chute and the scrotum anesthetized with local infiltration of 2% lidocaine hydrochloride. A vertical incision across the ventral aspect of the scrotum was made with a Newberry knife. The common vaginal tunic was incised, ligation of the vasculature was done with No. 2 chromic gut, and the testicle was removed with the accompanying adnexa. The epididymis and pampiniform plexus were removed before the measurements of testicular length and width were taken, as previously described. In addition to length and width, testicular volume and weight were measured by the water displacement method and a balance beam scale. All measurements were concluded within 10 rain of testicular excision. For testicular length and width from the caliper and ultrasound measurements, a mean value from the 3 independent measurements for each bull was calculated. The correlation coefficients and P values were then derived from these mean values and were compared with
Theriogenology
585
actual testicular length and width measurements after excision. Correlation coefficients were also calculated to test actual testicular volume and weight against the prolate spheroid formula for volume and its alteration for the prediction of weight obtained from measurements of length and width with both the caliper and ultrasound measurements in vivo. (Table 1) Table 1. Prediction of Testicle Volumes and Weights from Measurements Taken In Vivo for Length and Width with Calipers and Ultrasound. (Example demonstrates data from 2 bulls ) Bull Number 14 15 Bull Number
Actual Length __ 11.8 14.0
Actual Width 6.6 7.0
Actual Predicted Testicle Ultrasound Volume Volume 14 260 236.9 15 290 245.5 Predicted Volume = 0.5236(L)(W) "~ Predicted Weight = 0.5533(L)(W) 2
Ultrasound Length 11.4 11.1
Ultrasound Width 6.3 6.5
Caliper _Length 11.7 13.1
Caliper Width 6.9 7.5
Predicted Caliper Volume 291.7 385.8
Actual Testicle Weight 280 315
Predicted Ultrasound Weight 250.3 259.5
Predicted Caliper _W_eight 308.2 407.7 i
A length-to-width ratio was calculated for each individual testicle (n = 116) obtained from the slaughterhouse and from the post-excision trial. A subjective evaluation was made of the testicular shape prior to castration, with all bulls having a normal ovoid-shaped scrotal mass. RESULTS The reliability of the formula for a prolate spheroid to calculate testicular volume was tested and verified with excised testicles (n = 116). A correlation coefficient was calculated from actual testicular volume by the water displacement method for each testicle and the predicted volume from the formula for a prolate spheroid. The predictive value for calculating the volume from the formula of prolate spheroid is accurate and reliable, with a correlation coefficient (Figure 1) of rZ= 0.8928 (P < 0.05). A second set of linear regressions was determined for actual testicular volume and testicular weight (Figure 2). The significant correlation coefficient of r2= 0.9776 (P < 0.05) indicates a high degree of relationship between testicular volume and weight. Since volume and weight of the individual testicles were highly correlated, a calculated mathematical alteration in the formula for the volume of a prolate spheroid was tested for reliability in predicting testicular weight from measurements of both the length and width of individual testicles. A correlation coefficient (Figure 3) of r2= 0.9084 (P < 0.05) was achieved when testing predicted weight from the derived formula and actual testicular weight. The previously reported formula of V = (FI)(r2)(H) for calculating the volume of a cylinder used in the computation of testicular volume was also tested (7,8). A correlation coefficient of r 2 = 0.5364 (P < 0.05) was obtained using testicular length and width of excised testicles with the
Theriogenology
586
formula for cylinder volume. This indicates a more reliable (r2= 0.8928; P < 0.05) predictor using the prolate spheroid formula for the calculation of testicle volume. Caliper and ultrasonographic measurements obtained in vivo for testicular length and width were evaluated and verified. In vivo caliper measurements for length compared with the actual testicular length (Figure 4) revealed a correlation coefficient of r2= 0.8023 (P < 0.05). Y = 1.0961(X) - 7.5094
600.0
r 2 = 0.8928
500.0
_= O > 400.0 .~ 0)
~,
O
n = 116
6_ _ q ~ o
300.0
O ~ o
°°
m
400
500
ty-
200.0 100.0 0.0 0
100
200
300
Actual Testicle Volume (cc's) Figure 1. Linear regression and correlation coefficient for actual testicle volume compared with the predicted testicle volume obtained from the formula for a prolate spheroid.
~o
Y = 1.0725(X) + 4.4083
500
r 2 = 0.9776
400 O
>.
t~
_
~
300
•,~
200
E-~
100
<
n = 116
t~,~r°
j
J
0 0
_
I 100
I
I
I
200 300 400 Actual Testicle Weight (gms)
500
Figure 2. Linear regression and correlation coefficient for actual testicle weight compared with actual testicle volume.
Theriogonology
587 Y =10726(X) - 9.8923
600
r 2 = 0.9084 y:
500
n = 116
400 "U 300
¢ 2oo .~
100 0
,,r
,
0
I
F
I
100 200 300 400 Actual Testicle Weight (gms)
500
Figure 3. Linear regression and correlation coefficient for actual testicle weight compared with the predicted testicle weight obtained from the modified formula for a prolate spheroid.
Y = 0.9976(X) - 0.2817 r 2 = 0.8023
"~ 16 a~ 14 = 12 10 qD 8 '~ 6
¢ e~
n=')0
4 2 I
0
2
4 6 8 10 12 Actual Testicle Length (cms)
14
Figure 4. Linear regression and correlation coefficient for actual testicle length and caliper acquired testicle lengths.
This same instrument for width (Figure 5) demonstrated a slightly lower correlation coefficient of r2= 0.7313 (P < 0.05). Ultrasonographic lengths and widths suggest that width is much more
Theriogenology
588
reliable than length when taken with the ultrasound instrument, length correlation coefficients of r2= 0.5111 (P < 0.05) for the ultrasound images were calculated (Figure 6) and compared with actual testicular lengths. Width correlation coefficients (Figure 7), however, were slightly more reliable, with values of r2= 0.8310 (P < 0.05). This demonstrates the degree of difficulty in obtaining accurate length measurements because of limitations with the ultrasound unit reproducing the entire testicle length on the screen and the curvilinear shape of the testicle.
~ 8.0 Y = 0.7449(X) + 2.0937 ~ 6.0 [ r 2 = 0 " 7 3 1 3 ....
4.o 2.0 0.0
I 0
I
t
2 4 6 A c t u a l Testicle W i d t h (cms)
Figure 5. L i n e a r regression and correlation coefficient for actual testicle width and caliper derived testicle width.
,~o 12.0 -cA 10.0 ©
u
Y = 0.5274(X) + 4.4133 r 2 = 0.5111 n = 20
~
~
o
~
8.0 6.0 4.0 2.0 o.o
I
I
~
I
I
L
I
2
4
6
8
10
12
14
Actual Testicle Length (cms) Figure 6. Linear regression and correlation coefficient for actual testicle length compared to testicle lengths obtained by ultrasound.
Theriogenology
589
Y = 0.6707(X) - 1.9171 *~
8.0
r 2 = 0.8310
60 -~.~
4.0
~:
2.0
0
2 4 6 Actual Testicle Width (cms)
Figure 7. Linear regression and correlation coefficient for actual testicle width c o m p a r e d to ultrasound obtained widths. Caliper and ultrasonography measurements for both length and width from in vivo testicles were used with the formulas for prediction of testicular volume and weight. The data suggests a slightly higher degree of reliability was achieved when measurements were obtained with the calipers. A correlation coefficient (Figure 8) of r2= 0.5497 (P < 0.05) was determined when calculating predicted volume from caliper measurements in vivo.
Y = 0.990
450
400 350
[-., .-.. "o ,~
300
"
250
"-"
( X ) + 38.55
r 2 = 0.5497 n=20
o
200 ~, ~
150
>
100
,,.-,
=
f
50
I
0
0
I
I
100 200 300 Actual Testicle V o l u m e (ccs)
400
Figure 8. Linear regression and correlation coefficient for actual testicle volume versus predicted testicle v o l u m e for length and width measurements obtained by caliper measurements.
Thedogeno~gy
590
A lower correlation coefficient (Figure 9) of ta= 0.3276 (P < 0.05) was derived for predicted testicular volume from measurements when using ultrasonographically obtained lengths and widths of testicles in vivo. The correlation coefficients achieved for predicted testicular weight with the respective instruments, where r2= 0.6340 (P< 0.05) for the calipers (Figure 10) and r2= 0.6249 (P < 0.05) for the ultrasound unit (Figure 11), indicate both instruments have essentially the same degree of accuracy in predicting testicle weight. Y = 0.3915(X) + 120.5
o
-o
300
[-
250
.~ ~~
200
"~ " "
150
.~_
100
•=
I:= O
~
>
r2 = 0.3276 n=20 o
O
o
50
D
0
1 O0
200
300
400
Actual Testicle Volume (ccs) Figure 9. Linear regression and correlation coefficient for actual testicle volume compared to predicted testicle volume for length and width measurements obtained by ultrasound.
Y = 1.275(x) -33.91 r2 = 0 6340 _-_= 20
450
"- ~
400
[...., .~--.-
350 300
•~ ~:
200 150 100 50
z 250
,~
=~
O
~
-
f v v
/-
o 0
50
100 150 2 0 0 2 5 0 3 0 0 3 5 0 4 0 0 Actual Testicle Weight (gm)
Figure 10. Linear regression and correlation coefficient for actual testicle weight compared to predicted testicle weight for length and width measurements obtained caliper length and width measurements.
Thefiogeno~gy
591
Y = 0.550(X) + 85.48
400 t.3 •=r~ 350 300 ,.-. 250
r 2 = 0.6249 n = 20
lOO f
~:
50 o
J
0
50
~
L
I
I
I
100 150 200 250 300 Actual Testicle Weight (gms)
350
Figure 11. Linear regression and correlation coefficient for actual testicle weight compared to predicted testicle weight for length and width measurements obtained by ultrasound. We also calculated from both populations of bulls (n = 116) in this project, a postexcision ratio of testicular length to width at 1.81:1 (SEM = 0.01). These results were obtained from bulls which had, based on subjective evaluation, a normal ovoid-shaped scrotum and testicular mass. DISCUSSION The formula for a prolate spheroid and its modification provides a rapid and highly reliable method for estimating testicular volume and weight from testicular length and width measurements. The advantages of length and width measurements and application to a simple mathematical formula are readily apparent. This formula utilizes real physical measurements of individual testicutar length and width, unlike that of SC which measures only 1 dimension and encompasses both testicles simultaneously. The formula would be useful in following an animal through a long term course of study for an accurate assessment of single testicle volume or weight changes with time. Bulls are often presented with a variation in testicular shape (1,10,11). Frequently seen are the long slender testicles, rounded testicles, and normal ovoid scrotum (1,10,11). The physical measurement of length and width can now be considered in the calculation of total testicular volume and weight, rather than a single measurement of SC, or testicular width, which may not reflect the total volume or weight. Since individual testicles can be measured and the formula applied, variation in testicular shape among individual bulls is discounted as a potential problem in calculating total volume or weight of the 2 testicles. Previous reports of formulas and regression equations for calculating testicular volume from in vivo length and width measurements were given and tested (7,8). The formula of V = (FI)(r2)(H) for computing the volume of a cylinder is referenced and used in the calculation of testicular volume (7,8). The bovine testicle is not synonymous with the shape of a cylinder (12).
592
Theriogenology
However, a prolate spheroid more closely resembles the true shape of the testicle and would therefore be more appropriate for computation of testicular volume (12,17). It is also important to note that volume is a real physical measurement of the testis, subject only to measurement error (17). The data from the regression equation is linear over the entire range of predicted testicular volumes and weights calculated against the actual measurements. This would suggest that testicular measurements are accurate in predicting the volume of testicles at extreme lengths and widths encountered in adult bulls. Regression analysis demonstrated testicular volume and weight are highly correlated. Therefore, the prolate spheroid formula for predicting volume could be easily converted for accurate determination of testicular weight. This was determined from mean weights and volumes, and the calculation of a constant for the weight formula, as 0.5236 is for the volume formula, where V = 0.5236(L)(W) 2. A factor of 0.5533 was found to be valid for the weight formulation, where testicular weight = 0.5533(L)(W) 2 . Analysis showed that measurement of testicular widths and lengths were good predictors of testicular weight; therefore, weight could be determined from these physical dimensions. The data demonstrates the reliability of using caliper and ultrasonographic imaging for measurements of testicular length and width. Both instruments are quite reliable for width measurements; however, calipers are easier to use and demonstrate a higher degree of accuracy for measurement of testicular length. Further work is needed to alleviate the problems encountered determining testicular lengths by ultrasound. This data suggests that caliper measurements for determining the length and width and subsequent calculation of testicular volume and weight would be a reliable method for use in bulls for individual testicles lying outside the normal ovoid parameters. Our justification for this assumption is that 2 physical measurements are taken, rather than a single measurement of the SC or testicular diameter. This would include both the bulls with the more rounded, spheroid-shaped testicles and those with long, slender testicles (1,10,11). Caliper and ultrasonographic measurements for length and width of the testicles had an acceptable correlation in the calculation of testicular weight. However, ultrasound values for the measurement of length and width were not as well correlated with the actual testicular volume. Difficulty was encountered when measuring a curvilinear structure in a 2 dimensional plane and accurately determining length. The computer-assisted scale contained within the ultrasound unit had a constraint of approximately 12 cm; therefore, testicles greater than 12 centimeters in length would be incomplete from pole to pole on the screen. To try and circumvent this problem, a "stand- off" pad was placed between the probe and the scrotum; however, little advantage was gained in viewing the testicle in its entirety. A gridline pad was also applied between the scrotum and the probe to measure from the cranial pole of the testicle to a marked midline on the grid, then from this midline to the caudal pole. This was cumbersome and produced great difficulty in obtaining accurate measurements. We conclude that ultrasound is applicable in diagnostics, but it adds no advantage over that of calipers for measuring the length of the in vivo testicles except when only a width measurement is desired. In the estimation of total weight or volume, both testicles should be measured for length and width to alleviate error in calculation because of possible disparate size of the 2 testicles. However, if the testicles are of equal size, one can be measured and the formula product multiplied by 2 for total testicular volume or weight.
Theriogenology
593
We conclude that the formula for a prolate spheroid, utilizing measurements for the dimensions of length and width, would be more accurate in determining testicular volume and the revised formula for weight of the long, slender or more rounded testicle (1). This formulation relies on 2 measurements for calculation, rather than only the one of width utilized in previous studies for the formulation of predicted volume and weight. We submit a more accurate assessment would be obtained from 2 physical measurements, especially in bulls with a variation in testicular shape. However, more research is needed in bulls with this variation. The ratio of testicular length to width in mature bulls with normal ovoid-shaped testicles is 1.8:1. This data is based on excised testicles and on the derivation of length-to-width ratios. This would be useful in objectively determining the physical shape of the individual bovine testicle. Again, we evaluated only those bulls with a normal ovoid-shaped testicular mass; therefore, more data needs to be collected from bulls with a variation in testicular shape. REFERENCES 1. Bailey TL, Monke DR, Hudson RS, Wolfe DF, Carson RL, and Riddell G. Testicular shape and its relationship to sperm production in mature Holstein bulls. Theriogenology 1996, 46:881-887. 2. Cartee RE, Gray BW, Powe TA, Hudson RS, Whitesides J. Preliminary implications of BMode ultrasonography of the testicles of beef bulls with normal breeding soundness examinations. Theriogenology 1989;31:1149-1157. 3. Cartee RE, Powe TA, Gray BW, Hudson RS, Kuhlers DL. Ultrasonographic examination of the boar testicle. Am J Vet Res 1986;47:2543-2548. 4. Cates WF. Observations on Scrotal Circumference and Its Relationship to Classification of Bulls. Proc Ann Meeting of the Soc for Theriogenology 1975; 3-18. 5. Coulter GH. Puberty and postpubertal development of beef bulls, in Morrow DA (ed), Current Therapy in Theriogenology. Philadelphia: WB Saunders Company, 1986; 142-148. 6. Elmore RG, Bierschwal CJ, Youngquist RS. Scrotal circumference measurements in 764 beef bulls. Theriogenology 1976;6:485-494. 7. Fields MJ, Burns WC, Warnick AC. Age, season and breed effects on testicular volume and semen traits in young beef bulls. J Anim Sci 1979;48:1229-1334. 8. Hahn JR, Foot RH, Seidel GE. Testicular growth and related sperm output in dairy bulls. J Anim Sci 1969;29:41. 9. Lunstra DD, Echternkamp SE. Puberty in beef bulls: acrosome morphology and semen quality in bulls of different breeds. J Anim Sci 1982;55:638-648. 10. Monke D. Examination of the bovine scrotum, testicles, and epididymides. Part I. Comp Cont Ed Pract Vet 1987;9:F252-F258. 11. Monke D. Examination of the bovine scrotum, testicles, and epididymides. Part II Comp Cont Ed Pract Vet 1987;9:F277-F283. 12. Morrill WK, Selby SM, Johnson WG. Surfaces and Curves. In: Modem Analytical Geometry. Scranton, Pa: Haddon Craftsman, Inc., 1972;398-432. 13. SAS Introductory Guide for Personal Computers. Cary NC: SAS Institute Inc, 1985. 14. Society for Theriogenology: Manual for Breeding Soundness Examination of Bulls. 1983;12. 15. Thompson HA, Johnson WH. Scrotal size of yearlings sires and early calving in beef herds: Epidemiological investigation of possible causal pathways. Theriogenology 1995;43:12791287.
594
Theriogenology
16. Veeramachaneni DNR, Ott RS, Heath EH, McEntee K, Bolt D J, Hixon, JE. Pathophysiology of small testes in beef bulls; relationship between scrotal circumference, histopathologic features of testes and epididymis, seminal characteristics and endocrine profiles. Am J Vet Res 1986;47:1988-1999. 17.Watson-Whitmyre M, Stetson MH. A mathematical model for estimating paired testes weight from in situ testicular measurements in three species of hamster. Anat Rec 1985;213:473-476. 18.Willet EL, Ohms JI. Measurements of testicular size and its relationship to production of spermatozoa by bulls. J Dairy Sci 1957;40:1559-1569.