Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Aug 12.
Published in final edited form as: Am J Primatol. 2010 Jun;72(6):530–538. doi: 10.1002/ajp.20807

Validation of Multi-Detector Computed Tomography as a Non-Invasive Method for Measuring Ovarian Volume in Macaques (Macaca fascicularis)

Jeryl C Jones 1,*, Susan E Appt 2, Stephen R Werre 3, Joshua C Tan 4, Jay R Kaplan 2
PMCID: PMC3741054  NIHMSID: NIHMS487375  PMID: 20131358

Abstract

The purpose of this study was to validate low radiation dose, contrast-enhanced, multi-detector computed tomography (MDCT) as a non-invasive method for measuring ovarian volume in macaques. Computed tomography scans of four known-volume phantoms and nine mature female cynomolgus macaques were acquired using a previously described, low radiation dose scanning protocol, intravenous contrast enhancement, and a 32-slice MDCT scanner. Immediately following MDCT, ovaries were surgically removed and the ovarian weights were measured. The ovarian volumes were determined using water displacement. A veterinary radiologist who was unaware of actual volumes measured ovarian CT volumes three times, using a laptop computer, pen display tablet, hand-traced regions of interest, and free image analysis software. A statistician selected and performed all tests comparing the actual and CT data. Ovaries were successfully located in all MDCT scans. The iliac arteries and veins, uterus, fallopian tubes, cervix, ureters, urinary bladder, rectum, and colon were also consistently visualized. Large antral follicles were detected in six ovaries. Phantom mean CT volume was 0.702±SD 0.504 cc and the mean actual volume was 0.743±SD 0.526 cc. Ovary mean CT volume was 0.258±SD 0.159 cc and mean water displacement volume was 0.257±SD 0.145 cc. For phantoms, the mean coefficient of variation for CT volumes was 2.5%. For ovaries, the least squares mean coefficient of variation for CT volumes was 5.4%. The ovarian CT volume was significantly associated with actual ovarian volume (ICC coefficient 0.79, regression coefficient 0.5, P = 0.0006) and the actual ovarian weight (ICC coefficient 0.62, regression coefficient 0.6, P = 0.015). There was no association between the CT volume accuracy and mean ovarian CT density (degree of intravenous contrast enhancement), and there was no proportional or fixed bias in the CT volume measurements. Findings from this study indicate that MDCT is a valid non-invasive technique for measuring the ovarian volume in macaques.

Keywords: macaque, menopause, ovary volume, computed tomography, MDCT

INTRODUCTION

Macaque species are important animal models for studies of menopause and related diseases in women [Shively & Clarkson, 2009]. Macaque models have been used for research on ovarian cycle control, endometriosis, anovulatory infertility, osteoporosis, stress-related infertility, menopausal changes in physiology, and related metabolic disorders (diabetes, atherosclerosis) [Weinbauer et al., 2008]. Like other macaques, cynomolgus monkeys (Macaca fascicularis) have a close phylogenetic relationship to human beings (greater than 95% homology in DNA), with females showing considerable reproductive and cardiovascular similarities to women [Kavanagh et al., 2005]. Similarities include reproductive system morphology, endocrine system, hormone receptors, the control of unilateral single-egg ovulation, and the predominant delivery of single offspring [Buse et al., 2008]. In women, ovarian volume has been found to be an accurate indicator of ovarian senescence and menopause onset [Bastos et al., 2006; Erdem et al., 2003; Flaws et al., 2001; Giacobbe et al., 2004; Lass et al., 1997; Oppermann et al., 2003; Pavlik et al., 2000]. Fertility declines as the ovarian volume decreases, and women with mean ovarian volumes of <3ml respond poorly to ovulation induction [Lass et al., 1997]. In a study in which infertile women were found to have significantly smaller ovaries than fertile women, follicle-stimulating hormone (FSH) concentrations were not different between groups [Erdem et al., 2003]. This observation suggests that ovarian volume may be a more sensitive marker of ovarian aging than serum FSH.

The current standard method for measuring the ovarian volume in macaque models is water displacement (Archimedes principle). This technique requires ovariectomy and therefore precludes longitudinal studies of the effects of age, hormonal levels, and experimental treatments on ovarian size. The standard technique for measuring the ovarian volume in women is transvaginal ultrasonography and volume calculation using a prolate ellipsoid formula (bidimensional measurement multiplied by coronal long axis) [Erdem et al., 2003; Tepper et al., 1995]. Transvaginal ultrasonography is not currently feasible in macaques due to their small size. While transabdominal ultrasonography has been previously described as a method for measuring mature ovarian follicles in monkeys, [Morgan et al., 1987] complete visualization of outer ovary margins can be limited due to superimposition of pelvic bones, or gas and fecal material in bowel loops. In addition, the prolate ellipsoid method for volume calculation method assumes a true oval shape for the object. This assumption is not consistently true for monkey ovaries.

Multi-detector computed tomography (MDCT) has been established as a rapid, non-invasive technique for evaluating anatomy and diseases of the pelvic region in women [Bazot et al., 1999; Foshager & Walsh, 1994; Kumar et al., 2002; MacSweeney & King, 1994; Occhipinti et al., 1993; Rigsby & Siegel, 1994; Rozenblit et al., 2001; Saksouk & Johnson, 2004]. Multi-detector CT uses x-rays, an array of multiple detectors, and advanced computer processing to generate high-detail slice images of the body [Horton et al., 2002]. Advantages of MDCT include high-speed acquisition of data, low operator dependency, and the ability to visualize structures of interest without superimposition of overlying bone, gas, feces, or other obstacles. Digital MDCT image data can also be stored as a phenotypic repository and later used to evaluate additional structures and make measurements without having to rescan the subject. Computed tomographic volume measurement (CT volumetry) can be performed for any structure included in the scan field, either through the use of automated edge-calculation software or by hand-tracing the edges in sequential transverse images [Breiman et al., 1982]. A previous study in phantoms and human prostate tumor patients demonstrated that volumes of irregularly shaped structures are more accurate and repeatable when measured using circumscribed tracings of object margins (manual circumscription) vs. those measured using prolate ellipsoid formulas [Rkein et al., 2009].

In our previous study, we found that low radiation dose, intravenous contrast-enhanced MDCT was a feasible technique for characterizing morphology of reproductive structures in adult female macaques [Jones et al., 2007]. The radiation dose measured in the pelvic region during scanning was less than 0.7 rad, far below that previously reported to cause ovarian effects in monkey models. The ovaries, iliac arteries and veins, uterus, fallopian tubes, cervix, vagina, ureters, urinary bladder, rectum, and colon were consistently visualized. Overlying bone was not an impediment. Scans of the pelvic region could be acquired in 10–15 sec, using mild sedation. No significant breathing motion artifacts were detected. The objective for the study presented in this report was to validate low radiation dose contrast-enhanced MDCT as a non-invasive method for quantifying the ovarian volume in macaques.

METHODS

This study was conducted at Wake Forest University between 2006 and 2009. All animal procedures were in compliance with federal, state, and institutional guidelines; and were approved by the Wake Forest University Institutional Animal Care and Use Committee.

Phantoms

Three tuberculin syringes were filled with whole milk in volumes of 0.25, 0.50, and 0.75 ml, respectively. A plastic container was filled with mayonnaise and a glass bead with a known water displacement volume of 1. 5 cc was suspended in the center.

Monkeys

Nine female cynomolgus macaques (Macaca fascicularis) were used for this study. Monkeys ranged in age from 5 to 15 years of age. For two monkeys, one ovary was included in the analyses and, for the remaining seven monkeys, both ovaries were included (N =16 ovaries). All macaques were fed an isoflavone-free diet prepared at the Wake Forest University Primate Center laboratory.

MDCT Scanning Procedure

Scans of phantoms and monkeys were acquired using a previously described, low radiation dose scanning protocol [Jones et al., 2007] and a 32-slice MDCT scanner [Aquilion 32, Toshiba America Medical Systems, 2441 Michelle Dr. Tustin, CA]. Immediately after scanning, images were reviewed using an image analysis workstation to insure they were of diagnostic quality [TeraRecon AquariusNet Server, TeraRecon, Inc. San Mateo, CA]. Technical parameters for the scanning protocol were: 80 kVp, 100 mA, 32-row spiral acquisition, detail reconstruction algorithm, 0.5 mm slice thickness, 0.5 sec rotation time, and axial reconstruction plane. Phantoms were placed on the CT table and slices were acquired perpendicular to the long axes of containers. Monkeys were sedated using ketamine [Ketaset 100 mg/ ml, Fort Dodge Animal Health, Fort Dodge, IA] 15 mg/kg IM or general anesthesia using isoflurane [Isoflurane USB, Webster Veterinary, Sterling, MA] 5% initially and maintained @ 1.5%. When fully relaxed, monkeys were positioned supine on the CT table with the legs placed on a cylindrical positioning pad. An indwelling catheter was inserted into the saphenous vein and taped into place. Vertical and horizontal pilot scans were used to set boundaries for the scan field to include the ilial wings, pubic bone, and tuber ischii in each monkey. Scans of monkeys were acquired before and after intravenous injection of iodinated contrast medium [Iopromide 300 mgI/ml, 2 ml/kg, Ultravist, Berlex, Montville, NJ]. Contrast delay times ranged from 10 to 40 sec.

Ovarian Weight and Water Displacement Volume

Immediately following CT scanning, monkeys were transported to an adjacent surgical suite. An experienced veterinarian (S. E. A.) removed ovaries via laparotomy. Immediately following removal, the adnexal tissues were trimmed from ovaries and ovarian weights were measured using a precision analytical balance [Mettler AE 100, Mettler Toledo, Inc. Columbus, OH]. Ovarian volume was then determined using water displacement. A 10 ml graduated cylinder was filled with enough water so that, when the ovary was placed in the graduated cylinder, it would be fully submerged. The volume of the water in the graduated cylinder was recorded (V1) and then the ovary was placed in the graduated cylinder and the new volume recorded (V2). The water displacement volume of the ovary was calculated as V2–V1.

MDCT Volume Measurement

Data from all MDCT scans were archived on a secure research server by the Wake Forest University Center for Biomolecular Imaging. A board-certified veterinary radiologist (J. C. J.) accessed the server and downloaded image files via the Internet to a laptop computer [MacBook Pro 17 in, 2.6GHz Intel Core, 2GB memory, Apple Inc., Cupertino, CA 95014]. Volumes were measured using an interactive pen display tablet [Cintiq 21UX, Wacom Technology Corp., Vancouver, WA] and medical image analysis freeware [OsiriX Open-source Imaging Software, v. 3.3.2, OsiriX Foundation, Geneva]. Display settings for volume measurements were: slice thickness 1mm, zoom factor 3, window width 350, and window level set at the mean CT density value (Hounsfield units, HU) for the center slice of the phantom or ovary. Phantoms and ovaries were first located using the multi-planar reformatting tool (Fig. 1). Regions of interest (ROI) were hand-traced around the margins of each phantom and each ovary in each transverse image, using the software’s pencil tool (Figs. 2 and 3). The observer did not review actual volumes prior to drawing ROI’s. ROI were summed using the software’s tool for ROI volume calculation. Mean CT density in HU was recorded from the summed ROI’s. All CT volume and density measurements were repeated three times. Images demonstrating each measurement were captured and archived.

Fig. 1.

Fig. 1

Open in a new tab

Transverse (A), dorsal planar (B), oblique dorsal planar maximum intensity projection (C) and 3D volume-rendered images (D) demonstrating the CT appearance of the right ovary and surrounding anatomic landmarks in a female macaque. The left ovary was previously removed. C, cervix; CF, coxofemoral joint; Co, colon; F, fallopian tube; IAV, iliac artery and vein; O, ovary; SIJ, sacroiliac joint; UB, urinary bladder; UR, ureter; UT, uterus.

Fig. 2.

Fig. 2

Open in a new tab

Screen-capture computer display demonstrating the summed regions of interest method used for calculation of CT volume and mean CT density of the glass bead phantom. The top-left frame displays the three-dimensional image of the bead that was created by summing the ROI’s for all transverse slices. The calculated values for density (HU) and volume (cm3) are displayed in a white box below the three-dimensional image. The bottom left frame demonstrates a manual ROI tracing (blue dotted line) around the edge of the phantom. The right frame displays a dorsal planar CT image of the bead suspended in a plastic container filled with mayonnaise. A green reference line identifies the location of the transverse CT slice that is displayed in the bottom left frame. A white histogram graph in the bottom right frame displays the range of CT densities within the bead. (Color figures can be viewed in the online issue, which is available at www.interscience.Wiley.com)

Fig. 3.

Fig. 3

Open in a new tab

Screen-capture computer display demonstrating the summed regions of interest method used for calculation of CT volume and mean CT density of the right ovary in a female macaque. The top-left frame displays the three-dimensional image of the ovary that was created by summing the ROI’s for all transverse slices. The calculated values for density (HU) and volume (cm3) are displayed in a white box below the three-dimensional image. The bottom left frame demonstrates a manual ROI tracing around the edge of the ovary (dotted blue line). The right frame displays a dorsal planar CT image of the ovary and uterus. A green reference line identifies the location of the transverse CT slice that is displayed in the bottom left frame. A white histogram graph in the bottom right frame displays the range of CT densities within the ovary. (Color figures can be viewed in the online issue, which is available at www.interscience.Wiley.com)

Statistical Analysis

The following hypotheses were tested by a statistician (S. R. W.) using commercially available software [SAS version 9.2, Cary, NC (hypotheses 1–4) and MedCalc, Mariakerke, Belgium (hypothesis 5)].

Hypothesis 1

CT volumetry is a precise and accurate technique. To assess precision, a mean coefficient of variation (with a 95% confidence interval) was estimated for phantom data, whereas least squares mean coefficient of variation (with a 95% CI) was estimated for monkey ovary data. For each phantom or ovary, accuracy was computed as 100 × (1–{|actual volume–CT volume|/actual volume}). Subsequently, mean accuracy (with a 95% CI) was estimated for phantom data, whereas least squares mean (with a 95% CI) was estimated for ovary data.

Hypothesis 2

There is a significant association between the CT volume and water-displacement volume of the ovaries. Association between the CT volume and actual volume was evaluated using a scatter plot followed by intraclass (ovary within monkey) correlation (ICC) and mixed-model regression analysis. To obtain the ICC coefficient, a variance components model was fitted to the data with monkey and ovary-within-monkey as random effects. Subsequently, the ICC coefficient was calculated as (variance due to monkey+variance due to ovary)/(variance due to monkey+variance due to ovary+residual variance). The Regression model had actual volume as the outcome, CT volume as a linear predictor, and monkey as a random effect.

Hypothesis 3

There is a significant association between the CT volume and ovary weight. Association between the CT volume and ovary weight was evaluated using a scatter plot followed by ICC and mixed-model regression analysis.

Hypothesis 4

CT volume accuracy is not affected by mean ovarian CT density.

Association between the CT volume and mean ovarian CT density was evaluated using a scatter plot followed by ICC and mixed-model regression analysis.

Hypothesis 5

There is no proportional or fixed bias in the CT volume measurements. To assess the presence or absence of bias in the CT volume measurements, data were analyzed using Deming regression. Statistical significance was set to α=0.05. Data analyses were performed using standard statistical analysis software.

RESULTS

Subjective MDCT Findings

Ovaries were successfully located in all scans. The iliac arteries and veins, uterus, fallopian tubes, cervix, vagina, ureters, urinary bladder, rectum, and colon were also consistently visualized (Fig. 1). Large antral follicles were detected in six ovaries, appearing as focal areas of non-enhancement or ringenhancement within the ovarian stroma. Outer ovarian margins were well-defined for eight ovaries, moderately defined for two ovaries, and poorly defined for six ovaries. For five ovaries, margins were indented or compressed by adjacent structures such as the uterus, urinary bladder, colon, or pelvic bones.

Precision and Accuracy of CT Volumetry

Phantom mean CT volume was 0.702±SD 0.504 cc whereas the corresponding mean actual volume was 0.743±SD 0.526 cc, resulting in a mean coefficient of variation for CT volumes of 2.5% (95% CI 0.3–4.7). Ovary mean CT volume was 0.258±SD 0.159 cc while the corresponding mean water displacement volume was 0.257±SD 0.145 cc. For phantoms, the mean coefficient of variation for CT volumes was 2.5% (95% CI 0.3–4.7). For ovaries, the least squares mean coefficient of variation for CT volumes was 5.4% (95% CI 2.8–8.1). Upper limits of the two confidence intervals were less than the normally accepted level of 10%. The least squares mean accuracy of CT volumetry was 94.3% (95% CI 90.4–98.1) for phantoms and 64.5% (95% CI 51.0–77.9) for ovaries.

Association Between CT Ovarian Volume and Actual Ovarian Volume

Scatter plot analysis demonstrated a linear relationship between the CT and water displacement (actual) ovarian volumes (Fig. 4). The intraclass correlation (ICC) coefficient for ovary-within-monkey was 0.79. Mixed model regression analysis showed that the linear relationship between the CT ovarian volume and actual ovarian volume was statistically significant (regression coefficient 0.50, 95% CI 0.3–0.7) (P =0.0006).

Fig. 4.

Fig. 4

Open in a new tab

Scatter plot demonstrating a linear relationship between ovary CT volume vs. actual volume by water displacement.

Association Between CT Volume and Ovary Weight

Scatter plot analysis demonstrated a linear relationship between the CT ovary volume and ovary weight. The ICC coefficient was 0.62. Mixed model regression analysis showed that the linear relationship between the CT volume and ovary weight was statistically significant (regression coefficient 0.60, 95% CI 0.2–1.0) (P = 0.015).

Effect of Ovarian Mean CT Density on CT Volume Accuracy

Scatter plot analysis did not reveal a linear relationship between the ovarian mean CT density and CT volume accuracy. The ICC coefficient was 0.00, and the regression coefficient was 0.1 (P=0.4518). This indicates that the degree of ovarian contrast-enhancement did not have a significant effect on CT volume accuracy.

Tests for Bias in CT Volume Measurements

Deming regression showed that the linear relationship between the actual ovary volume and CT ovary volume can be modeled as actual volume = −0.06+1.24 CT volume (i.e., Deming regression equation with an intercept of −0.06 and a regression coefficient of 1.24) (Fig. 5). The 95% confidence interval for the intercept (−0.28 to 0.15) included 0. This finding indicates that CT volume measurements did not have a fixed bias. Furthermore, the 95% confidence interval for the slope (0.24 to 2.23) included 1. This finding indicates CT volume measurements did not have a proportional bias.

Fig. 5.

Fig. 5

Open in a new tab

Plot demonstrating no evidence of bias for actual volume vs. CT volume measurements. The solid line represents the Deming regression equation [Actual volume = −0.06204+ (1.2358 × CT volume)], whereas the broken line represents the ordinary regression equation [x = y].

DISCUSSION

Conclusions from our study were: (1) CT volumetry is a precise and accurate technique for measuring ovarian volume in macaques, (2) The CT ovarian volume is significantly associated with the actual ovarian volume, (3) The CT ovarian volume is significantly associated with the actual ovarian weight, (4) there is no association between the CT volume accuracy and mean ovarian CT density (degree of contrast-enhancement), and (5) there is no proportional or fixed bias in CT volume measurements. Our study design and findings are consistent with previous studies validating CT volumetry as a technique for measuring the volumes of phantoms [Breiman et al., 1982; Brenner et al., 1982; Disler et al., 1994; Forbes et al., 1985; Nawaratne et al., 1997; Sohaib et al., 2000; Van Hoe et al., 1997; Wheatley et al., 1995]; the canine liver, spleen, and kidney [Breiman et al., 1982; Moss et al., 1981]; the porcine kidney [Cai et al., 2007; Chul Kim, 2004]; and the human kidney, liver, spleen, and brain [Brenner et al., 1982; Nawaratne et al., 1997; Wheatley et al., 1995].

Precision is the degree to which a result is reproduced each time the measurement is repeated [Lamb, 2008]. An established method for describing precision of a diagnostic test is to calculate the coefficient of variation, either within a single observer or between different observers [Stewart et al., 2008]. In our study, a single observer used manually drawn ROI’s to calculate phantom and ovary CT volumes. Precision was very high, with a measured intraobserver variation of 2.5% for phantom volumes and 5.4% for ovary CT volumes. This finding indicates that, for future studies, manual CT ovarian volume measurements are likely to be repeatable when performed by a single observer.

Accuracy is a measure of how closely a diagnostic test represents what it is intended to represent [Lamb, 2008]. Previous studies have found that the accuracy of the CT volume measurements can be reduced by systematic errors arising from the observer, relative density between the object of interest and adjacent structures, scanning technique, or image display setting [Brenner et al., 1982; Disler et al., 1994; Van Hoe et al., 1997]. In our study, phantom scans were used to detect and minimize technical sources of error as much as possible. For monkey scans, intravenous contrast was used to maximize the detection of ovarian margins vs. adnexa. A high-resolution display screen and pen drawing tool were used to minimize observer sources of error. Using a previously published accuracy formula for diagnostic imaging tests [Lamb, 2008] and water displacement volume as the gold standard, we calculated a high percentage accuracy value for CT volumes of phantoms and a moderate percentage accuracy value for CT volumes of ovaries (94 and 65%, respectively). The moderate accuracy value calculated for ovary CT volumes was based on paired comparisons for individual ovaries within monkeys. For some ovaries, including those described as having well-defined margins, the CT volumes and actual volumes were markedly different. The reason for this discrepancy is unknown, but the possibility of an error in recording of water displacement volumes cannot be excluded. When mean ovary CT volumes and mean actual volumes for the group of monkeys were calculated, the difference between these two values was only 0.001 cc. This finding indicates that, for future studies comparing longitudinal treatment effects among groups of monkeys, mean ovary CT volumes are likely to be representative of mean actual ovary volumes.

Associations between findings from a diagnostic test and findings from a gold standard measure are further evidence of the validity of a diagnostic test. We chose to use ICC rather than Pearson’s correlation for our comparisons because ovary-within-monkey was the unit of analysis [Farrell et al., 2001; Yen & Lo, 2002]. Although monkeys are independent, the left and right ovaries within a given monkey are not independent. For Pearson’s correlation to be valid, all units of analysis have to be independent. In our study, ICC and linear regression tests were used to demonstrate that the CT volume estimates for ovaries were significantly associated with the ovarian water displacement volumes and ovarian weights.

Tests of association between CT accuracy and ovarian density were performed to determine whether variation in the degree of ovarian contrast-enhancement was a significant factor for CT volume accuracy. In our study, iodinated contrast medium was administered intravenously and scans were initiated within 40 sec of contrast administration. This protocol was based on findings from our previous study indicating that ovarian tissue exhibited a different degree of contrast-enhancement than adjacent adnexa and that this made ovarian margins more distinguishable for ROI tracing. Previous studies have found that difficulties in identifying and tracing margins of an object can be important sources of error in CT volume measurements [Breiman et al., 1982; Brenner et al., 1982]. We attempted to standardize the rate of contrast administration and the contrast delay time as much as possible, but some variations in the degree of ovarian contrast-enhancement occurred in spite of our efforts. However, we found no significant association between CT ovarian volume accuracy and ovarian density (i.e., degree of contrast-enhancement). This finding indicates that, for future studies, variations in ovarian contrast-enhancement over time should not have a significant effect on CT volume accuracy.

Another measure of the validity of a diagnostic test is whether or not there is evidence of bias in the measurements [Nawaratne et al., 1997]. Bias is an average deviation from a true value [Johnson, 2008]. Systematic errors (bias) can be either fixed (a difference that is constant across the range of values) or proportional (difference that increases or decreases across a range of values) [Lamb, 2008]. In our study, we used Deming regression to assess bias in CT volumetry for predicting actual ovarian volume [Martin, 2000]. This test is better than ordinary regression analysis when the independent (CT volume) and dependent (actual ovarian volume) variables tend to vary randomly. We found no evidence of either fixed or proportional bias in our CT volume measurements. This finding further supports the validity of ovarian CT volumetry as a diagnostic test for use in future studies.

Limitations of our study design included a small sample size, lack of multiple observers for CT volume measurements, and lack of repeated measurements or multiple observers for water displacement ovarian volume measurements. The small sample size may have limited the power of our tests of association. Having one observer perform all CT volume measurements precluded calculations of reliability. Reliability of a diagnostic test is a measure of how often the same result would be recorded by different observers. Having water displacement volumes for ovaries measured once and by one observer precluded calculation of both precision and reliability for that test. Water displacement volumes were also recorded to one decimal place, vs. three decimal places recorded for CT ovary volumes. It is therefore possible that the gold standard water displacement volumes may have been a source of error in some of our individual CT ovary volume accuracy calculations. Ovary weight and histologic ovary volume may be worth investigating as alternative gold standards for future studies. Ovary weight can be easily recorded to three decimal places using a precision balance and we found that this measure was also significantly associated with CT ovarian volume in our study. Histologic volume was used as the gold standard for measuring CT volume accuracy in a previous study tracking ovarian maturation in female [Muller et al., 2004].

In summary, findings from the study described in this report validate low-radiation dose contrast-enhanced MDCT as non-invasive method for measuring the ovarian volume in female macaques. Scans were acquired in a few seconds, data were securely archived and retrieved for later analyses without having to rescan the animal, and ovaries were identified in all scans. Other pelvic canal structures such as the iliac arteries and veins, ureters, urinary bladder, uterus, fallopian tubes, cervix, colon, and rectum were also well-visualized. The ovary volume measurements could be performed using a commercially available laptop computer, pen display tablet, and free image analysis software. Variations in the degree of ovarian contrast-enhancement did not have a significant effect on CT volume accuracy. Future studies are needed to determine the sensitivity of MDCT for detecting reproductive diseases and ovarian volume longitudinal changes in macaque models. Ovarian weight and histologic volume may also warrant further investigation as alternative gold standard methods for measuring the ovarian volume in macaques.

ACKNOWLEDGMENTS

All animal procedures were in accordance with institutional animal care regulations and applicable national laws.

Contract grant sponsor: National Institutes of Health; Contract grant number: R24 RR 022191.

REFERENCES

  1. Bastos CA, Oppermann K, Fuchs SC, Donato GB, Spritzer PM. Determinants of ovarian volume in pre-, menopausal transition, and post-menopausal women: a population-based study. Maturitas. 2006;53:405–412. doi: 10.1016/j.maturitas.2005.07.002. [DOI] [PubMed] [Google Scholar]
  2. Bazot M, Deligne L, Boudghene F, Buy JN, Lassau JP, Bigot JM. Correlation between computed tomography and gross anatomy of the suspensory ligament of the ovary. Surgical and Radiologic Anatomy. 1999;21:341–346. doi: 10.1007/BF01631336. [DOI] [PubMed] [Google Scholar]
  3. Breiman RS, Beck JW, Korobkin M, Glenny R, Akwari OE, Heaston DK, Moore AV, Ram PC. Volume determinations using computed tomography. AJR American Journal of Roentgenology. 1982;138:329–333. doi: 10.2214/ajr.138.2.329. [DOI] [PubMed] [Google Scholar]
  4. Brenner DE, Whitley NO, Houk TL, Aisner J, Wiernik P, Whitley J. Volume determinations in computed tomography. Journal of American Medical Association. 1982;247:1299–1302. [PubMed] [Google Scholar]
  5. Buse E, Zoller M, Van Esch E. The macaque ovary, with special reference to the cynomolgus macaque (Macaca fascicularis) Toxicologic Pathology. 2008;36:24S–66S. [Google Scholar]
  6. Cai W, Holalkere NS, Harris G, Sahani D, Yoshida H. Dynamic-threshold level set method for volumetry of porcine kidney in CT images in vivo and ex vivo assessment of the accuracy of volume measurement. Academic Radiology. 2007;14:890–896. doi: 10.1016/j.acra.2007.03.005. [DOI] [PubMed] [Google Scholar]
  7. Chul J. Animal study of renal volume measurement on abdominal CT using digital image processing; preliminary report. Clinical Imaging. 2004;28:135–137. doi: 10.1016/S0899-7071(03)00115-3. [DOI] [PubMed] [Google Scholar]
  8. Disler DG, Marr DS, Rosenthal DI. Accuracy of volume measurements of computed tomography and magnetic resonance imaging phantoms by three-dimensional reconstruction and preliminary clinical application. Investigative Radiology. 1994;29:739–745. doi: 10.1097/00004424-199408000-00002. [DOI] [PubMed] [Google Scholar]
  9. Erdem M, Erdem A, Biberoglu K, Arslan M. Age-related changes in ovarian volume, antral follicle counts and basal follicle stimulating hormone levels: comparison between fertile and infertile women. Gynecological Endocrinology. 2003;17:199–205. doi: 10.1080/gye.17.3.199.205. [DOI] [PubMed] [Google Scholar]
  10. Farrell T, Leslie JR, Chien PF, Agustsson P. The reliability and validity of three dimensional ultrasound volumetric measurements using an in vitro balloon and in vivo uterine model. British Journal of Obstetrics and Gynaecology. 2001;108:573–582. doi: 10.1111/j.1471-0528.2001.00148.x. [DOI] [PubMed] [Google Scholar]
  11. Flaws JA, Langenberg P, Babus JK, Hirshfield AN, Sharara FI. Ovarian volume and antral follicle counts as indicators of menopausal status. Menopause. 2001;8:175–180. doi: 10.1097/00042192-200105000-00006. [DOI] [PubMed] [Google Scholar]
  12. Forbes G, Gehring DG, Gorman CA, Brennan MD, Jackson IT. Volume measurements of normal orbital structures by computed tomographic analysis. AJR American Journal of Roentgenology. 1985;145:149–154. doi: 10.2214/ajr.145.1.149. [DOI] [PubMed] [Google Scholar]
  13. Foshager MC, Walsh JW. CT anatomy of the female pelvis: a second look. Radiographics. 1994;14:51–64. doi: 10.1148/radiographics.14.1.8128066. discussion 64–66. [DOI] [PubMed] [Google Scholar]
  14. Giacobbe M, Mendes Pinto-Neto A, Simoes Costa-Paiva LH, Martinez EZ. The usefulness of ovarian volume, antral follicle count and age as predictors of menopausal status. Climacteric. 2004;7:255–260. doi: 10.1080/13697130410001713715. [DOI] [PubMed] [Google Scholar]
  15. Horton KM, Sheth S, Corl F, Fishman EK. Multidetector row CT: principles and clinical applications. Critical Reviews in Computed Tomography. 2002;43:143–181. [PubMed] [Google Scholar]
  16. Johnson R. Assessment of bias with emphasis on method comparison. The Clinical Biochemist Reviews. 2008;29:S37–S42. [PMC free article] [PubMed] [Google Scholar]
  17. Jones JC, Appt SE, Bourland JD, Hoyer PB, Clarkson TB, Kaplan JR. Multidetector computed tomographic morphology of ovaries in cynomolgus macaques (Macaca fascicularis) Journal of American Association for Laboratory Animal Science. 2007;46:54–63. [PubMed] [Google Scholar]
  18. Kavanagh K, Koudy Williams J, Wagner JD. Naturally occurring menopause in cynomolgus monkeys: changes in hormone, lipid, and carbohydrate measures with hormonal status. Journal of Medical Primatology. 2005;34:171–177. doi: 10.1111/j.1600-0684.2005.00114.x. [DOI] [PubMed] [Google Scholar]
  19. Kumar P, Rehani MM, Kumar L, Sharma R, Bhatla N, Singh R, Sundaram KR. Potential of CT-scan based tumor volume as a response indicator in chemotherapy of advanced epithelial ovarian cancer. Medical Science Monitor. 2002;8:CR667–CR674. [PubMed] [Google Scholar]
  20. Lamb CR. Statistical briefing: accuracy and precision. Veterinary Radiology and Ultrasound. 2008;49:600–601. doi: 10.1111/j.1740-8261.2008.00438.x. [DOI] [PubMed] [Google Scholar]
  21. Lass A, Silye R, Abrams DC, Krausz T, Hovatta O, Margara R, Winston RM. Follicular density in ovarian biopsy of infertile women: a novel method to assess ovarian reserve. Human Reproduction. 1997;12:1028–1031. doi: 10.1093/humrep/12.5.1028. [DOI] [PubMed] [Google Scholar]
  22. MacSweeney JE, King DM. Computed tomography, diagnosis, staging and follow-up of pure granulosa cell tumour of the ovary. Clinical Radiology. 1994;49:241–245. doi: 10.1016/s0009-9260(05)81848-3. [DOI] [PubMed] [Google Scholar]
  23. Martin RF. General deming regression for estimating systematic bias and its confidence interval in methodcomparison studies. Clinical Chemistry. 2000;46:100–104. [PubMed] [Google Scholar]
  24. Morgan PM, Hutz RJ, Kraus EM, Cormie JA, Dierschke DJ, Bavister BD. Evaluation of ultrasonography for monitoring follicular growth in rhesus monkeys. Theriogenology. 1987;27:769–780. doi: 10.1016/0093-691x(87)90299-8. [DOI] [PubMed] [Google Scholar]
  25. Moss AA, Friedman MA, Brito AC. Determination of liver, kidney, and spleen volumes by computed tomography: an experimental study in dogs. Journal of Computer Assisted Tomography. 1981;5:12–14. doi: 10.1097/00004728-198102000-00002. [DOI] [PubMed] [Google Scholar]
  26. Muller T, Molnar T, Szabo A, Romvari R, Hancz C, Bercsenyi M, Horn P. Tracking of the hormonally induced maturation of female eels by computed tomography. Acta Veterinaria Hungarica. 2004;52:235–243. doi: 10.1556/AVet.52.2004.2.12. [DOI] [PubMed] [Google Scholar]
  27. Nawaratne S, Fabiny R, Brien JE, Zalcberg J, Cosolo W, Whan A, Morgan DJ. Accuracy of volume measurement using helical CT. Journal of Computer Assisted Tomography. 1997;21:481–486. doi: 10.1097/00004728-199705000-00027. [DOI] [PubMed] [Google Scholar]
  28. Occhipinti KA, Frankel SD, Hricak H. The ovary. Computed tomography and magnetic resonance imaging. Radiologic Clinics of North America. 1993;31:1115–1132. [PubMed] [Google Scholar]
  29. Oppermann K, Fuchs SC, Spritzer PM. Ovarian volume in pre- and perimenopausal women: a population-based study. Menopause. 2003;10:209–213. doi: 10.1097/00042192-200310030-00006. [DOI] [PubMed] [Google Scholar]
  30. Pavlik EJ, DePriest PD, Gallion HH, Ueland FR, Reedy MB, Kryscio RJ, van Nagell JR., Jr Ovarian volume related to age. Gynecologic Oncology. 2000;77:410–412. doi: 10.1006/gyno.2000.5783. [DOI] [PubMed] [Google Scholar]
  31. Rigsby CK, Siegel MJ. CT appearance of pediatric ovaries and uterus. Journal of Computer Assisted Tomography. 1994;18:72–76. doi: 10.1097/00004728-199401000-00016. [DOI] [PubMed] [Google Scholar]
  32. Rkein AM, Harrigal C, Friedman AC, Persky D, Krupinski E. Comparison of the accuracy of CT volume calculated by circumscription to prolate ellipsoid volume (bidimensional measurement multiplied by coronal long axis) Academic Radiology. 2009;16:181–186. doi: 10.1016/j.acra.2008.08.013. [DOI] [PubMed] [Google Scholar]
  33. Rozenblit AM, Ricci ZJ, Tuvia J, Amis ES., Jr Incompetent and dilated ovarian veins: a common CT finding in asymptomatic parous women. AJR American Journal of Roentgenology. 2001;176:119–122. doi: 10.2214/ajr.176.1.1760119. [DOI] [PubMed] [Google Scholar]
  34. Saksouk FA, Johnson SC. Recognition of the ovaries and ovarian origin of pelvic masses with CT. Radiographics. 2004;24:S133–S146. doi: 10.1148/rg.24si045507. [DOI] [PubMed] [Google Scholar]
  35. Shively CA, Clarkson TB. The unique value of primate models in translational research. Nonhuman primate models of women’s health: introduction and overview. American Journal of Primatology. 2009;71:715–721. doi: 10.1002/ajp.20720. [DOI] [PubMed] [Google Scholar]
  36. Sohaib SA, Turner B, Hanson JA, Farquharson M, Oliver RT, Reznek RH. CT assessment of tumour response to treatment: comparison of linear, cross-sectional and volumetric measures of tumour size. The British Journal of Radiology. 2000;73:1178–1184. doi: 10.1259/bjr.73.875.11144795. [DOI] [PubMed] [Google Scholar]
  37. Stewart EE, Chen X, Hadway J, Lee TY. Hepatic perfusion in a tumor model using DCE-CT: an accuracy and precision study. Physics in Medicine and Biology. 2008;53:4249–4267. doi: 10.1088/0031-9155/53/16/003. [DOI] [PubMed] [Google Scholar]
  38. Tepper R, Zalel Y, Markov S, Cohen I, Beyth Y. Ovarian volume in postmenopausal women—suggestions to an ovarian size nomogram for menopausal age. Acta Obstetricia et Gynecologica Scandinavica. 1995;74:208–211. doi: 10.3109/00016349509008940. [DOI] [PubMed] [Google Scholar]
  39. Van Hoe L, Haven F, Bellon E, Baert AL, Bosmans H, Feron M, Suetens P, Marchal G. Factors influencing the accuracy of volume measurements in spiral CT: a phantom study. Journal of Computer Assisted Tomography. 1997;21:332–338. doi: 10.1097/00004728-199703000-00034. [DOI] [PubMed] [Google Scholar]
  40. Weinbauer GF, Niehoff M, Neihaus M, Srivastav S, Fuchs A, Van Esch E, Cline JM. Physiology and endocrinology of the ovarian cycle in Macaques. Toxicologic Pathology. 2008;36:7S–23S. doi: 10.1177/0192623308327412. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Wheatley JM, Rosenfield NS, Heller G, Feldstein D, LaQuaglia MP. Validation of a technique of computer-aided tumor volume determination. Journal of Surgical Research. 1995;59:621–626. doi: 10.1006/jsre.1995.1214. [DOI] [PubMed] [Google Scholar]
  42. Yen M, Lo LH. Examining test-retest reliability: an intraclass correlation approach. Nursing Research. 2002;51:59–62. doi: 10.1097/00006199-200201000-00009. [DOI] [PubMed] [Google Scholar]

RESOURCES