- Email: [email protected]
Ultrasonographic and laparoscopic evaluation of the reproductive tract of the captive female African lion (Panthera leo)
Available online at www.sciencedirect.com
Theriogenology 76 (2011) 810 – 818 www.theriojournal.com
Ultrasonographic and laparoscopic evaluation of the reproductive tract of the captive female African lion (Panthera leo) Robert M. Kirbergera,*,1, Martin L. Schulmanb,1, Marthinus J. Hartmana a
Department of Companion Animal Clinical Studies, Faculty of Veterinary Science, University of Pretoria, Private Bag X04, Onderstepoort, 0110 South Africa b Department of Production Animal Studies, Faculty of Veterinary Science, University of Pretoria, Private Bag X04, Onderstepoort, 0110 South Africa Received 10 December 2010; received in revised form 4 April 2011; accepted 8 April 2011
Abstract The use of transabdominal ultrasonography to assess the oestrous cycle has not been previously described in the African lion (Panthera leo). Twelve sexually mature lionesses and five female cubs had their reproductive organs assessed by transabdominal ultrasound. Ovarian findings were compared to laparoscopic findings while performing laparoscopic ovariectomy or salpingectomy. Vaginal cytology was performed and serum progesterone levels were determined. By combining all data the oestrous cycle stage of each lion was determined. One lion was far pregnant and was not operated on. In adults a uterine body could be seen ultrasonographically in 67% of lions while mural structures could be distinguished in 44% of lions. Five uterine horns could be seen in 3 lions. In 12 adults 10 ovaries were found of which eight had discernable follicles or luteal structures. During laparoscopy 12 active ovaries were seen with luteal structures seen in 11 ovaries and follicles in 2 ovaries. Using laparoscopy as the gold standard, ultrasonography had a sensitivity of 66% and specificity of 83% to detect ovarian reproductive activity. Two uterine cysts and a cluster of periovarian cysts were seen in three different lions. Three lions were pregnant, two were in oestrus, three in a luteal phase (dioestrus), and four were in anoestrus. Transabdominal ultrasound in combination with serum progesterone levels and vaginal cytology can be used to assess ovarian cyclical activity with reasonable accuracy in captive bred lions. © 2011 Elsevier Inc. All rights reserved. Keywords: Cyclical activity; Laparoscopy; Lioness; Panthera leo; Reproduction; Ultrasound
1. Introduction The last 10 years have seen marked changes in the reproductive management of wild carnivores, including lions, either free ranging or in captivity. This is mainly due to the increasing availability, technological advances, and improved operator expertise in trans-abdominal or trans1
R.M. Kirberger and M.L. Schulman made an equal contribution to the article. * Corresponding author. Tel ⫹27 12 5298270; fax: ⫹27 12 5298307. E-mail address: [email protected] (R. Kirberger). 0093-691X/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.theriogenology.2011.04.013
rectal ultrasonography as well as improved endocrine analysis and reproductive cycle control measures. In the female African lion (Panthera leo), behavioural observations, clinical, and laboratory methods have all been reported for the purpose of oestrous monitoring and defining the reproductive status of an individual animal [1,2]. A few reports describe the application of vaginal cytology [1,2] and more frequently reported are hormonal methods utilising either faecal or serum samples [1– 6]. These methods have been reported in both free-ranging and captive lion populations. The application of diagnostic ultrasound has been
R.M. Kirberger et al. / Theriogenology 76 (2011) 810 – 818
previously reported mainly for diagnosis of pregnancy in both free-ranging and captive lionesses. The method provides a non-invasive and relatively uncomplicated means to assess reproductive status. There are reports of both trans-rectal and trans-abdominal approaches, primarily with the aim of visualization of the gravid uterus [2]. Few descriptions of the normal reproductive tract have been reported, particularly with regard to imaging of the ovaries. The ability to observe changes in ovarian function is limited to an extent by the small size of the ovarian structures in domestic felids [5]. The utilization of ovarian imaging as an adjunct to oestrous monitoring has, however, been reported in both domestic and non-domestic felids [5,7–9]. Ultrasonography can potentially provide a rapid alternative to laboratorybased methods for the reproductive evaluation of wild felids by direct visualization of the dynamic changes during both the follicular and luteal phases of the oestrous cycle as well as for the diagnosis (and staging) of pregnancy and pathology of the reproductive tract. This is particularly useful for in-the-field application, a reality of studying both free-ranging and captive wild felids. The four phases of the felid oestrous cycle are prooestrus, oestrus, dioestrus, and anoestrus. Pro-oestrus and oestrus are both associated with the sequential development of visible follicles in the ovaries. Thereafter, during dioestrus, one or more corpora luteae (CL) are visible, often with a visible transition in development from a corpus haemorrhagicum (CH). Anoestrus (interoestrus) is the period of follicular quiescence with visible non-functional follicular and luteal sites [1,3]. Vaginal cytology relies on the effects of oestrogenisation resulting in an increased proportion of cornified vaginal epithelial cells and a clear background on fixed and stained vaginal epithelium smears to describe prooestrus and oestrus [1,2]. Hormonal methods describing the oestrous cycle depend on steroids from either serum or faecal samples. The cyclical elevations of oestrogens differentiate anoestrus from oestrus periods. Currently we also realise that the lioness will exhibit both induced ovulation post-mating prior to the luteal phase, as well as spontaneous ovulation, with associated elevations in progestagens after a surge in oestrogens [1,3,4]. The luteal phase, in both pregnant and non-pregnant domestic cats is associated with qualitatively similar progestagen concentrations, with a shorter non-pregnant luteal phase [3,4]. The pregnant luteal phase in the African lion may be associated with a greater elevation in mean serum progesterone concentrations [2,6]. The objectives of this study were, first, to assess the
811
usefulness of trans-abdominal ultrasound observations of the reproductive tract of 17 captive-bred lionesses by comparison with laparoscopic visualisation at the time of sterilisation. These methods, together with the data derived from concurrent measurement of serum progesterone concentration (SPC) and vaginal cytology, were used to define the reproductive status of the lionesses. 2. Materials and methods 2.1. Animals Lions were raised as captive lions in the Limpopo province of South Africa, 130 km NW of Pretoria. They were housed from being solitary to up to 12 lions in various sized camps ranging from 162 to 720 m2 in a total fenced enclosure of 138 ⫻ 296 m. They were fed a diet of about 30 kg of 10% fresh chicken and 90% bovine and various species of game meat every 5 days. The University of Pretoria was requested to implement a surgical sterilization programme to limit population growth. The project was approved by the University’s Animal Use and Care Committee and Research Committee (protocol number V051-10). Twelve sexually mature female lions of varying parity with a median age of 9 years (range 3–10 years) and a median weight of 138 kg (range 100 –165 kg) were examined during the spring of 2010. The ages of the adult lionesses were provided by the owner and were approximate ages and the cubs were aged according to their dentition [10]. Three were classified as pregnant based on external appearance with enlarged mammae and typical ultrasonographic and laparoscopic features. One of these was anaesthetized and examined but was not operated on due to potential risks associated with her advanced pregnancy status. An additional five female cubs aged about 8 months old with a median weight of 42 kg (range 39 – 45) were also examined. All lionesses were examined on one occasion over a five-day observation period under general anaesthesia. They were immobilised with a dart gun and combinations of tiletamine and zolazapam (Zoletil Virbac, Halfway House, South Africa) and medetomidine (Domitor, Pfizer Animal Health, Sandton, South Africa). Propofol (Diprivan, Fresenius Kabi, Midrand, South Africa) was used to facilitate intubation. Lions were maintained on isoflurane (Isofor, Safeline Pharmaceuticals, Johannesburg, South Africa). After the procedure the anaesthesia was antagonised with atipamazole (Antisedan, Pfizer Animal Health, Sandton, South Africa). Adult lions were fasted for 72 h and cubs 24 h preoperatively, but had free access to water.
812
R.M. Kirberger et al. / Theriogenology 76 (2011) 810 – 818
tended). The terminal descending colon content and any effects on visualizing reproductive structures was recorded. As abdominal fat hampers ultrasonographic examination an obesity index (OI) was devised by measuring the subcutaneous fat from the skin to peritoneum at the umbilicus and ventral intra-abdominal fat layer 2 cm cranial to the umbilicus parasagittaly to the right. These two measurements were combined and divided by the body weight to give the OI (the higher the index the more abdominal obesity was present), e.g., an OI of 0.10 would be very thin and an OI of 0.40 very obese.
The lionesses were part of a laparoscopic surgical sterilization trial comparing ovariectomy to salpyngectomy that will be reported elsewhere. 2.2. Ultrasonography Each lioness underwent a dorsally recumbent transabdominal ultrasound examination of the reproductive tract. A portable ultrasound machine was used with a 2– 6 MHz convex array transducer for the adults, and a 3.5–7.5 MHz microconvex array transducer (Aloka Prosound 2, AXIM, Midrand, South Africa) for the cubs. The uterine body and horns and both ovaries were examined and their associated size, visible length, echogenicity, wall layering, luminal content, and presence of structures (including follicles, CH, and CL) were recorded. Length was determined from an optimal sagittal and width from an optimal transverse plane through the structure evaluated. All visualized round structures ⱖ 5.0 mm in diameter were classified as follicles if they were anechoic and had acoustic enhancement; CH if slightly echogenic with minimal acoustic enhancement; and CL if slightly hyperechoic to the surrounding ovarian tissue with no acoustic enhancement [7]. The location of the ovaries relative to the ipsilateral kidney and surrounding structures were recorded. The ovarian size and structure was compared to direct visualization during the ensuing laparoscopic procedure. Three lionesses were pregnant. Their ovaries were examined as above. The bladder size in each animal was evaluated and given a value from 0 (empty) to 4 (markedly dis-
2.3. Laparoscopy Each lioness (n ⫽ 16) underwent laparoscopy upon completion of the ultrasonographic examination starting in dorsal recumbency. The lionesses were then tilted 45° to the left, and 45° to the right, successively, and the right and left ovaries and associated structures, respectively, were identified and digital images captured. The authors examined the images retrospectively and described the observed ovarian structures, follicles, or luteal structure (either CL or CH) and compared them to the ultrasonographic images obtained. 2.4. Reproductive status 2.4.1. Hormone assays Serum samples were collected for SPC determination by radioimmune assay (RIA). The samples were stored at ⫺40 °C until thawing for assay using proges-
Table 1 Serum progesterone concentration (SPC) in nmol/l, vaginal cytology superficial cell index (SCI%), reproductive status and ultrasonographic visibility of uterine body and horns, and number of ovarian structures seen on ultrasound compared to laparoscopic findings. Lion
Uterus
Left ovary
Seen
Follicle no.
CH no.
Right ovary CL no.
Follicle no.
CH no.
CL no.
No.
SPC
SCI%
Cycle
Body
LH
RH
Seen
US
Lap
US
Lap
US
Lap
Seen
US
Lap
US
Lap
US
Lap
1 2 3 4 5 6 7 8 9 10 11 12
7.5 1.6 196.3 198.1 140 ND 139.4 1.2 164.1 3.6 82.4 1.4
0 0 0 0 0 0 20 100 80 0 20 0
An An P Di P P Di O O An Di An
Y Y Y N Y Y Y Y N Y Y N
N N Y Y Y Y Y Y N Y Y N
N N Y Y Y Y N Y Y N N N
N N ND Y N N N Y Y Y N Y
— — — 1 — — — — — — — —
— — — 2 ND — — — — — — —
— — — — — — — 1 1 — — —
— — — — ND — — 0 3 — — —
— — — — — — — — — — — —
— — — — ND — 2 — — — 2 —
Y N Y Y Y N N Y Y N N N
— — — — ND — — 3 — — — —
— — — — ND — — 3 — — — —
— — — — ND — — — 1 — — —
— — — — ND — — — 1 — — —
— — — — ND — — — — — — —
— — — — ND 1 1 — — — 2 —
Key: An, anoestrus; Di, dioestrus; O, oestrus; P, pregnant; Y, yes; N, no; US, ultrasound; Lap, laparoscopy; CH no., corpus haemorrhagicum number; CL no., corpus luteum number; ND, not done due to advanced pregnancy.
R.M. Kirberger et al. / Theriogenology 76 (2011) 810 – 818
813
terone RIA kits (Progesterone Coat-a-Count, Diagnostic Products) as previously described [11]. 2.4.2. Vaginal cytology Vaginal smears were collected from each of the 12 sexually mature animals immediately prior to ultrasonography. These were obtained after passage of a sterilized, clear pyrex tube speculum of 12 mm diameter via the vulva and vestibulum. Thereafter, a sterile cotton-tipped swab attached to a stainless steel wire was passed through the tube until making contact with the cranial vaginal mucous membrane. The swab was rotated several times before retrieval and then rolled gently on a glass microscope slide, air-dried, and ® stained with Diff-Quik (Kyroquick, Kyron Laboratories, Benrose, South® Africa) before permanent mounting using Entellan (Merck, Halfway House, South Africa) and a glass cover slip. The slide was examined microscopically to describe the presence and percentage of superficial epithelial cells expressed as the Superficial Cell Index (SCI), and the clarity of the background as an indicator of oestrogenisation [1,2]. 2.4.3. Ovarian structures The ovaries of adult animals were observed by a combination of ultrasonography and laparoscopy in situ. The ultrasonographic variables recorded were ovarian dimensions (length, width, and height) and a prolate ellipsoid ovarian volume calculated using the method of Pavlik et al [12] for each observed ovary. All observed structures ⬎ 5.0 mm in diameter (follicles, CH, and CL) were recorded (Table 1). The ovaries were imaged on laparoscopic examination in all animals. Ovarian activity was defined by the observation of prominent structures. The presence of clear, vesicular appearing structures ⬎ 0.50 mm was recorded as follicles. The luteal phase was defined by observation of one or more prominent, somewhat irregular yellowish structures ⬎ 0.50 mm [13]. Based on the combination of observed ultrasonographic and laparoscopic ovarian structures, vaginal cytology, and the SPC, the non-pregnant lionesses were classified as being in oestrus (pro-oestrus), dioestrus, or anoestrus [7]. 3. Results 3.1. Ultrasound In the nine non-pregnant adult lionesses, six animals (66.7%) had visible uterine bodies dorsal to the bladder (Table 1). Of these, four (44.4%) had wall layering visible with the mean ventral and dorsal wall thickness being 3.3
Fig. 1. Uterine body ultrasonographic images (Lion 10): (A) sagittal image with near wall 4 mm and far wall 3.8 mm as indicated by cursors. Anechoic bladder in near field and hyperechoic line of ventral colonic wall just beyond uterus; (B) transverse image with total diameter of 10 mm between cursors. Bladder in near field and curved hyperechoic line of colonic wall ventrally and to right of uterus.
and 3.7 mm respectively (Fig. 1). Two of the uteri had anechoic content up to 1.7 mm wide and only one had a central echogenic line indicative of the wall-lumen interface but with no content present. Total mean uterine wall diameter was 7.9 mm (range 5.8 –10) and mean visible length was 43.9 mm. A separate vagina, cervix and uterine body could not be distinguished in any lioness. Terminal colonic gas interfered with visualizing the uterine body in one lioness. In only one lioness could both uterine horns be traced cranially, and then only for a short distance. These were both 7 mm wide and had a mixed echogenicity. In one lioness, both uterine horns and in another lioness one uterine horn could be traced caudally from the ovaries for up to 48 mm and in only one of these was wall layering distinct due to a small amount of anechoic seg-
814
R.M. Kirberger et al. / Theriogenology 76 (2011) 810 – 818
Table 2 Dimensions of 10* ovaries seen on ultrasound.
Length (mm) Width (mm) Height (mm) Volume (cm3)
Median
Min
Max
27.47 17.8 17.87 4.6
19 9.5 11.2 1.1
40 28.3 30.6 18.1
* Eleven ovaries were seen but only 10 measured.
mented fluid in the horn. This horn measured 10.1 mm in diameter whereas the other two horns were 6 and 6.6 mm in diameter, respectively. In the cubs, a uterine body could only be found in one cub where its diameter was 4.2 mm and it was visible for 30 mm dorsal to the bladder. No uterine horns could be seen. In the 12 adult lionesses, only 11 ovaries were located
and described of which in three lionesses both ovaries could be seen (Table 1). The observed ovarian dimensions are shown in Table 2. Eight ovaries had an irregular outline, each of which had at least a single obvious structure, i.e. follicle or luteal structure (CL or CH) protruding from the surface (Figs. 2–4). Two ovaries had a smooth outline of which one had a central 7.5 mm CH or CL whereas the other ovary had no obvious ultrasonographic evidence of cyclical activity. Follicles were seen on three ovaries, one of which was a very large structure with a diameter of 27.9 mm (Fig. 3), Structures classified as CH’s were seen on four ovaries and CL’s on two ovaries. The smallest observed structure was 6 mm in diameter. On the left side, colonic gas and on the right side, small intestines, respectively occasionally interfered with attempts to image the ovaries.
Fig. 2. Ultrasonographic (A and C) and laparoscopic images (B and D) (Lion 4): (A) left ovary 17.5 mm long diagnosed as having 1 follicle (F) caudally; (B) Left ovary diagnosed as having 2 follicles (F) located more caudally; (C) right ovary diagnosed as having 1 CH and several smaller unrecognized structures. Note the longitudinal image of a segment of small intestine in the near field (D) right ovary diagnosed as having 5 CH (3 larger ones annotated). Note the small periuterine cyst (arrow).
R.M. Kirberger et al. / Theriogenology 76 (2011) 810 – 818
815
and ovaries could still be seen while in several of the thinner lions some structures were still not visible. In cubs, the OI was much higher (median 0.34; range 0.22– 0.71 and no reproductive structures could be seen. 3.2. Laparoscopy
Fig. 3. Ultrasonographic and laparoscopic images (Lion 8): (A) right ovary (30.6 ⫻ 28.3 ⫻ 16.8 mm) diagnosed as having 1 large (27.9 mm diameter) and 2 small follicles (not in this image); (B) right ovary diagnosed as having 1 large follicle cranially.
In adults, the ovary was often located in the region where the caudal hyperechoic edge of the caudal perirenal fat joined the ventral hyperechoic surface edge of the iliopsoas muscle at an angle of about 60 degrees (Fig. 5). In adults the left ovary was on average 44.6 mm directly caudal to the left kidney and 50 mm caudal to the right kidney. In the cubs no ovaries were seen and the anatomical landmarks described above were poorly defined. Only three bladders were moderately filled, with the remainder having less urine volumes and mean bladder filling was 1.4. Data showed that the extent of bladder filling did not seem to influence visibility of the uterine body. The OI in adults ranged from 0.21 to 0.42 (median 0.27) and did not appear to influence the visibility of structures. In some obese lionesses the uterine body
Seven lions had a total of 12 active ovaries containing luteal structures in 10 ovaries and follicles in two ovaries (Table 1). One lion had two follicles on the left ovary and five CLs on the right ovary with a SPC of 94.5 nmol/l and no apparent signs of oestrogenisation on vaginal cytology (Lion 4). Two lions had small single periuterine cysts (Figs. 2D and 4B) and one lion had a cluster of periovarian cysts (⬎ 10) adjacent to each ovary (Fig. 6). Comparing ultrasonographic and laparoscopic ovarian findings in 11 lions that had undergone both procedures (Table 1), there was no evidence of cyclic ovarian activity (presence of follicles or luteal structures) in nine ovaries of six lions and two of these ovaries were found on ultrasound. Five lions had visible signs of reproductive activity (follicles or luteal structures) in six ovaries on both ultrasonography and laparoscopy and these modalities agreed 100% on the identification of the type of activity. Six ovaries of four lions had reproductive activity observed on laparoscopy but not on ultrasound and in two ovaries of two lions ultrasound detected reproductive activity with no activity found on laparoscopy (Table 1). Using laparoscopy as the gold standard to detect follicles and luteal structures, ultrasound had a sensitivity of 66% and specificity of 83% adult lionesses. 3.3. Reproductive status The SPC, vaginal cytology findings, and reproductive status are given in Table 1. Reproductive status of the mature lionesses (n ⫽ 12) was determined from a combination of data derived from ultrasonographic and laparoscopic observed ovarian activity, SPC, and vaginal cytology. In the prepubertal cubs (n ⫽ 5) SPC was a mean of 3.12 (range 2.9 –3.4) giving baseline reproductive activity values. Three lionesses were pregnant. Of the nine nonpregnant animals, two (22.2%) were in oestrus (follicular phase), three (33.3%) in the luteal phase (dioestrus), and four (44.4%) in anoestrus (Table 1).
816
R.M. Kirberger et al. / Theriogenology 76 (2011) 810 – 818
Fig. 4. Ultrasonographic (A and C) and laparoscopic images (B and D) (Lion 9): (A) left ovary (21.9 ⫻ 19 ⫻ 20.7 mm) diagnosed as having a single CH; (B) Left ovary diagnosed as having 3 CH (2 visible on this image). Note the small periuterine cyst (arrow); (C) right ovary (24 ⫻ 15.5 ⫻ 15.2 mm) diagnosed as having a single CH; (D) Right ovary diagnosed as having 1 CH. Bowel grasper and LigaSure® hand piece used for salpyngectomy.
4. Discussion This study was enhanced by direct laparoscopic visualisation to provide both a sensitive and specific comparative standard for concurrent ultrasonographic observations. This was reflected in the decreased sensitivity and specificity to detect follicles and luteal structures ultrasonographically as compared to laparoscopy. A further advantage of laparoscopic examination of the felid ovary is that the latter does not have an enveloping ovarian bursa (as in the Canidae), and this, together with manipulation of the ovary and its associated structures and adnexa, enables complete visualization of the ovary. The normal ovary without the presence of any distinct associated structures was difficult to visualize ultrasonographically and was only seen on the left side, where there is less small intestines, in two
lions. Similarly, the normal, non-gravid and particularly anoestrus uterus was difficult to visualize via ultrasound as previously reported [7]. In our reproductive clinic small animals are routinely examined in the standing position to allow the uterus to fall to the dependent surface to enhance visibility. This is impractical in the lion under anaesthesia and dorsal recumbency may thus exacerbate the difficulty in visualising the reproductive tract. Ultrasonographic detection of ovulation and hence definition of oestrous cycle status may be controversial to establish with great certainty on ultrasound alone [7,9]. However the sensitivity of 66% to detect follicles and luteal structures ultrasonographically in mature adult lions does imply that there is a place for ultrasound in evaluating the reproductive tract. Ultrasonography identified follicles on three ovaries, with laparoscopy
R.M. Kirberger et al. / Theriogenology 76 (2011) 810 – 818
Fig. 5. Ultrasonographic image of left ovary indicating location landmarks. Cursors indicate caudal perirenal fat with the caudal pole of the kidney indicated by short arrows. The long arrows are in the iliopsoas muscle with the tips at the ventral border of the muscle.
showing follicles on only two ovaries. It is assumed that ovulation did not take place in the interval between the completion of the ultrasound examination and the laparoscopic visualization of the ovary, which took about 30 min. A distinct ultrasonographic differentiation of the luteal structures (CL versus CH) was observed. Including the pregnant lions, six luteal structures were seen on six ovaries of five lions, i.e., never more than a single luteal structure per ovary, whereas laparoscopy showed 22 luteal structures on 10 ovaries in seven lions, i.e., an average of 2.2 per ovary. The OI was developed to assess if obesity played a role in hampering visibility of the reproductive structures, which it did not appear to do in this trial. Fat results in increased attenuation of the ultrasound beam and loss of image quality. However, in this study the presence of follicles or luteal structures appeared to be the most important factor to enhance ovary visibility. Operator experience (in this case a board certified radiologist) and good quality ultrasound equipment also enhance the ability to detect ovaries. The laparoscopic observation of the entire ovarian surface and the readily-identifiable luteal structures allows accurate calculation of an ovulation rate. This ovulation rate is dependent on the number of follicles successfully undergoing maturation, ovulation, and luteinization, and excludes anovulatory follicles. This observation in conjunction with observed litter sizes provides an improved index of both reproductive efficiency and fertility in this polytocous species as the
817
maximum litter size is dependent on the ovulation rate. The result is a ratio between litter size and ovulation rate in a particular oestrous cycle, rather than a categorical birth rate that suffices in monotocous species [13]. The number of CL’s observed suggest that on average 3.6 ovulations per cycle occurred in this population of pregnant and non-pregnant captive female lions. This is similar to the reported litter sizes in free-ranging populations in Southern Africa, where average litter sizes reported are 3.02 and 3.1, respectively, and a maximum of six cubs [14]. This observation implies the ovulation, fertilization, and litter rates, respectively, may be similar in this species as a whole. In the non-pregnant lions in this study the ovulations were presumably induced, albeit without actual coitus having occurred as these females were never in direct physical contact with any male lions, but male lions were certainly within close proximity in nearby enclosures. This observation shows a relatively high ovulation rate occurred without coitus in this population. There may obviously be a discrepancy in free-living populations where only data describing litter rates are reported, however, early neonatal attrition rates may be a confounding factor here. The SPC values must be interpreted in the light of variable endocrine results in immobilized or anaesthetized wild felids due to potential effects of these procedures on hormonal secretion [6]. This is also compounded by the observation of unexplained “pul-
Fig. 6. Laparoscopic image of a right ovary undergoing electrocoagulation of the uterine artery for ovariectomy. Note cluster of periovarian cysts caudoventrally to the ovary.
818
R.M. Kirberger et al. / Theriogenology 76 (2011) 810 – 818
sations” in circulatory SPC in lions [6]. The observation of active luteal structures via SPC, ultrasonography, and laparoscopy in three females without direct physical access to any adult male supports previous reports of spontaneous ovulation as a feature of the oestrous cycle in lions [1,3,4]. Laparoscopy showed that two lions had small single periuterine cysts and one lion had a cluster of periovarian cysts (⬎ 10) adjacent to each ovary. These were non-pregnant lions, giving a prevalence of 33% in non-pregnant lions in this study population. The significance of these structures is, however, uncertain and future laparoscopic studies may shed more light on their clinical relevance. In conclusion, transabdominal ultrasonography in combination with serum progesterone levels and vaginal cytology can be used to assess reproductive activity with some accuracy in captive bred lions. These procedures may offer potential benefits in assessment of individual reproductive function, including subfertility, and assisted reproduction techniques in similar populations of lions.
[2]
[3] [4] [5] [6]
[7] [8]
[9]
[10] [11]
Acknowledgments The authors thank E. Monnet, L. Venter, L. Bester, R. Serfontein, R. Fourie, and students who were involved with the ultrasonographic and laparoscopic procedures, as well as Karl Storz for sponsoring laparoscopic equipment. References [1] Schmidt AM, Nadal LA, Schmidt MJ, Beamer NB. Serum concentrations of estradiol and progesterone during normal oes-
[12]
[13]
[14]
trous cycle and early pregnancy in the lion (Panthero leo). J Reprod Ferti 1979;57:267–72. Bertschinger HJ, de Barros Vaz Guimaraes MA, Trigg TE, Human, A. The use of deslorelin implants for the long-term contraception of lionesses and tigers Wildlife Res 2008;35: 525–30. Brown JL. Female reproductive cycles of wild female felids. Anim Repr Sc 2010; doi:10.1016/Anireprosci.2010.08.024. Brown JL. Comparative endocrinology of domestic and nondomestic felids. Theriogenology 2006;6:25–36. Goodrowe KL, Feline reproduction and artificial breeding technologies. Jnl Repro Sci 1992;28:389 –97. Brown JL, Bush M, Packer C, Pusey AE, Montfort SL, O’Brien JL, Janssen DL, Wildt DE. Hormonal characteristics of free ranging female lions (Panthera leo) of the Serengeti plains and Ngorongoro crater. J Reprod Fert 1993;87:107–14. Davidson AP, Baker TW. Reproductive ultrasound of the bitch and queen. Top Comp An Med 2009;24:55– 63. Peter AT, Jakovljevic S. Real-time ultrasonography of the small animal reproductive organs. Comp Contin Educ Pract Vet 1992; 14:739 – 44. Root CR, Spaulding KA. Diagnostic imaging in companion animal theriogenology. Seminars in Vet Med and Surg (Small Animal) 1994;9:7–27. Smuts GL, Anderson JL, Austin JC. Age determination of the African lion 1978;185:115⫺46. Bertschinger HJ, Asa CS, Calle PP, Long JA, Bauman K, DeMatteo K, Jöchle W, Trigg TE, Human A. Control of reproduction and sex related behaviour in exotic wild carnivores with the GnRH analogue deslorelin: preliminary observations. J Reprod Fertil Suppl 2001;57:275– 83. Pavlik EJ, DePriest PD, Gallion HH, Ueland FR, Reedy MB, Kryscio RJ, van Nagell JR. Ovarian volume related to age. Gyn Oncol 2000;77:410 –2. Nothling JO, de Cramer KGM, Gerber D, Kammer VR. Luteal and follicular count in bitches: Assessment by means of magnetic resonance imaging. Theriogenology 2006;66:1343–54. Mills MGL, Bester MN. In: Mammals of the Southern African Subregion, Skinner JD and Chimimba CT (Eds) 3rd ed., Cambridge University Press, Cambridge; 2005, p. 396.
Theriogenology 76 (2011) 810 – 818 www.theriojournal.com
Ultrasonographic and laparoscopic evaluation of the reproductive tract of the captive female African lion (Panthera leo) Robert M. Kirbergera,*,1, Martin L. Schulmanb,1, Marthinus J. Hartmana a
Department of Companion Animal Clinical Studies, Faculty of Veterinary Science, University of Pretoria, Private Bag X04, Onderstepoort, 0110 South Africa b Department of Production Animal Studies, Faculty of Veterinary Science, University of Pretoria, Private Bag X04, Onderstepoort, 0110 South Africa Received 10 December 2010; received in revised form 4 April 2011; accepted 8 April 2011
Abstract The use of transabdominal ultrasonography to assess the oestrous cycle has not been previously described in the African lion (Panthera leo). Twelve sexually mature lionesses and five female cubs had their reproductive organs assessed by transabdominal ultrasound. Ovarian findings were compared to laparoscopic findings while performing laparoscopic ovariectomy or salpingectomy. Vaginal cytology was performed and serum progesterone levels were determined. By combining all data the oestrous cycle stage of each lion was determined. One lion was far pregnant and was not operated on. In adults a uterine body could be seen ultrasonographically in 67% of lions while mural structures could be distinguished in 44% of lions. Five uterine horns could be seen in 3 lions. In 12 adults 10 ovaries were found of which eight had discernable follicles or luteal structures. During laparoscopy 12 active ovaries were seen with luteal structures seen in 11 ovaries and follicles in 2 ovaries. Using laparoscopy as the gold standard, ultrasonography had a sensitivity of 66% and specificity of 83% to detect ovarian reproductive activity. Two uterine cysts and a cluster of periovarian cysts were seen in three different lions. Three lions were pregnant, two were in oestrus, three in a luteal phase (dioestrus), and four were in anoestrus. Transabdominal ultrasound in combination with serum progesterone levels and vaginal cytology can be used to assess ovarian cyclical activity with reasonable accuracy in captive bred lions. © 2011 Elsevier Inc. All rights reserved. Keywords: Cyclical activity; Laparoscopy; Lioness; Panthera leo; Reproduction; Ultrasound
1. Introduction The last 10 years have seen marked changes in the reproductive management of wild carnivores, including lions, either free ranging or in captivity. This is mainly due to the increasing availability, technological advances, and improved operator expertise in trans-abdominal or trans1
R.M. Kirberger and M.L. Schulman made an equal contribution to the article. * Corresponding author. Tel ⫹27 12 5298270; fax: ⫹27 12 5298307. E-mail address: [email protected] (R. Kirberger). 0093-691X/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.theriogenology.2011.04.013
rectal ultrasonography as well as improved endocrine analysis and reproductive cycle control measures. In the female African lion (Panthera leo), behavioural observations, clinical, and laboratory methods have all been reported for the purpose of oestrous monitoring and defining the reproductive status of an individual animal [1,2]. A few reports describe the application of vaginal cytology [1,2] and more frequently reported are hormonal methods utilising either faecal or serum samples [1– 6]. These methods have been reported in both free-ranging and captive lion populations. The application of diagnostic ultrasound has been
R.M. Kirberger et al. / Theriogenology 76 (2011) 810 – 818
previously reported mainly for diagnosis of pregnancy in both free-ranging and captive lionesses. The method provides a non-invasive and relatively uncomplicated means to assess reproductive status. There are reports of both trans-rectal and trans-abdominal approaches, primarily with the aim of visualization of the gravid uterus [2]. Few descriptions of the normal reproductive tract have been reported, particularly with regard to imaging of the ovaries. The ability to observe changes in ovarian function is limited to an extent by the small size of the ovarian structures in domestic felids [5]. The utilization of ovarian imaging as an adjunct to oestrous monitoring has, however, been reported in both domestic and non-domestic felids [5,7–9]. Ultrasonography can potentially provide a rapid alternative to laboratorybased methods for the reproductive evaluation of wild felids by direct visualization of the dynamic changes during both the follicular and luteal phases of the oestrous cycle as well as for the diagnosis (and staging) of pregnancy and pathology of the reproductive tract. This is particularly useful for in-the-field application, a reality of studying both free-ranging and captive wild felids. The four phases of the felid oestrous cycle are prooestrus, oestrus, dioestrus, and anoestrus. Pro-oestrus and oestrus are both associated with the sequential development of visible follicles in the ovaries. Thereafter, during dioestrus, one or more corpora luteae (CL) are visible, often with a visible transition in development from a corpus haemorrhagicum (CH). Anoestrus (interoestrus) is the period of follicular quiescence with visible non-functional follicular and luteal sites [1,3]. Vaginal cytology relies on the effects of oestrogenisation resulting in an increased proportion of cornified vaginal epithelial cells and a clear background on fixed and stained vaginal epithelium smears to describe prooestrus and oestrus [1,2]. Hormonal methods describing the oestrous cycle depend on steroids from either serum or faecal samples. The cyclical elevations of oestrogens differentiate anoestrus from oestrus periods. Currently we also realise that the lioness will exhibit both induced ovulation post-mating prior to the luteal phase, as well as spontaneous ovulation, with associated elevations in progestagens after a surge in oestrogens [1,3,4]. The luteal phase, in both pregnant and non-pregnant domestic cats is associated with qualitatively similar progestagen concentrations, with a shorter non-pregnant luteal phase [3,4]. The pregnant luteal phase in the African lion may be associated with a greater elevation in mean serum progesterone concentrations [2,6]. The objectives of this study were, first, to assess the
811
usefulness of trans-abdominal ultrasound observations of the reproductive tract of 17 captive-bred lionesses by comparison with laparoscopic visualisation at the time of sterilisation. These methods, together with the data derived from concurrent measurement of serum progesterone concentration (SPC) and vaginal cytology, were used to define the reproductive status of the lionesses. 2. Materials and methods 2.1. Animals Lions were raised as captive lions in the Limpopo province of South Africa, 130 km NW of Pretoria. They were housed from being solitary to up to 12 lions in various sized camps ranging from 162 to 720 m2 in a total fenced enclosure of 138 ⫻ 296 m. They were fed a diet of about 30 kg of 10% fresh chicken and 90% bovine and various species of game meat every 5 days. The University of Pretoria was requested to implement a surgical sterilization programme to limit population growth. The project was approved by the University’s Animal Use and Care Committee and Research Committee (protocol number V051-10). Twelve sexually mature female lions of varying parity with a median age of 9 years (range 3–10 years) and a median weight of 138 kg (range 100 –165 kg) were examined during the spring of 2010. The ages of the adult lionesses were provided by the owner and were approximate ages and the cubs were aged according to their dentition [10]. Three were classified as pregnant based on external appearance with enlarged mammae and typical ultrasonographic and laparoscopic features. One of these was anaesthetized and examined but was not operated on due to potential risks associated with her advanced pregnancy status. An additional five female cubs aged about 8 months old with a median weight of 42 kg (range 39 – 45) were also examined. All lionesses were examined on one occasion over a five-day observation period under general anaesthesia. They were immobilised with a dart gun and combinations of tiletamine and zolazapam (Zoletil Virbac, Halfway House, South Africa) and medetomidine (Domitor, Pfizer Animal Health, Sandton, South Africa). Propofol (Diprivan, Fresenius Kabi, Midrand, South Africa) was used to facilitate intubation. Lions were maintained on isoflurane (Isofor, Safeline Pharmaceuticals, Johannesburg, South Africa). After the procedure the anaesthesia was antagonised with atipamazole (Antisedan, Pfizer Animal Health, Sandton, South Africa). Adult lions were fasted for 72 h and cubs 24 h preoperatively, but had free access to water.
812
R.M. Kirberger et al. / Theriogenology 76 (2011) 810 – 818
tended). The terminal descending colon content and any effects on visualizing reproductive structures was recorded. As abdominal fat hampers ultrasonographic examination an obesity index (OI) was devised by measuring the subcutaneous fat from the skin to peritoneum at the umbilicus and ventral intra-abdominal fat layer 2 cm cranial to the umbilicus parasagittaly to the right. These two measurements were combined and divided by the body weight to give the OI (the higher the index the more abdominal obesity was present), e.g., an OI of 0.10 would be very thin and an OI of 0.40 very obese.
The lionesses were part of a laparoscopic surgical sterilization trial comparing ovariectomy to salpyngectomy that will be reported elsewhere. 2.2. Ultrasonography Each lioness underwent a dorsally recumbent transabdominal ultrasound examination of the reproductive tract. A portable ultrasound machine was used with a 2– 6 MHz convex array transducer for the adults, and a 3.5–7.5 MHz microconvex array transducer (Aloka Prosound 2, AXIM, Midrand, South Africa) for the cubs. The uterine body and horns and both ovaries were examined and their associated size, visible length, echogenicity, wall layering, luminal content, and presence of structures (including follicles, CH, and CL) were recorded. Length was determined from an optimal sagittal and width from an optimal transverse plane through the structure evaluated. All visualized round structures ⱖ 5.0 mm in diameter were classified as follicles if they were anechoic and had acoustic enhancement; CH if slightly echogenic with minimal acoustic enhancement; and CL if slightly hyperechoic to the surrounding ovarian tissue with no acoustic enhancement [7]. The location of the ovaries relative to the ipsilateral kidney and surrounding structures were recorded. The ovarian size and structure was compared to direct visualization during the ensuing laparoscopic procedure. Three lionesses were pregnant. Their ovaries were examined as above. The bladder size in each animal was evaluated and given a value from 0 (empty) to 4 (markedly dis-
2.3. Laparoscopy Each lioness (n ⫽ 16) underwent laparoscopy upon completion of the ultrasonographic examination starting in dorsal recumbency. The lionesses were then tilted 45° to the left, and 45° to the right, successively, and the right and left ovaries and associated structures, respectively, were identified and digital images captured. The authors examined the images retrospectively and described the observed ovarian structures, follicles, or luteal structure (either CL or CH) and compared them to the ultrasonographic images obtained. 2.4. Reproductive status 2.4.1. Hormone assays Serum samples were collected for SPC determination by radioimmune assay (RIA). The samples were stored at ⫺40 °C until thawing for assay using proges-
Table 1 Serum progesterone concentration (SPC) in nmol/l, vaginal cytology superficial cell index (SCI%), reproductive status and ultrasonographic visibility of uterine body and horns, and number of ovarian structures seen on ultrasound compared to laparoscopic findings. Lion
Uterus
Left ovary
Seen
Follicle no.
CH no.
Right ovary CL no.
Follicle no.
CH no.
CL no.
No.
SPC
SCI%
Cycle
Body
LH
RH
Seen
US
Lap
US
Lap
US
Lap
Seen
US
Lap
US
Lap
US
Lap
1 2 3 4 5 6 7 8 9 10 11 12
7.5 1.6 196.3 198.1 140 ND 139.4 1.2 164.1 3.6 82.4 1.4
0 0 0 0 0 0 20 100 80 0 20 0
An An P Di P P Di O O An Di An
Y Y Y N Y Y Y Y N Y Y N
N N Y Y Y Y Y Y N Y Y N
N N Y Y Y Y N Y Y N N N
N N ND Y N N N Y Y Y N Y
— — — 1 — — — — — — — —
— — — 2 ND — — — — — — —
— — — — — — — 1 1 — — —
— — — — ND — — 0 3 — — —
— — — — — — — — — — — —
— — — — ND — 2 — — — 2 —
Y N Y Y Y N N Y Y N N N
— — — — ND — — 3 — — — —
— — — — ND — — 3 — — — —
— — — — ND — — — 1 — — —
— — — — ND — — — 1 — — —
— — — — ND — — — — — — —
— — — — ND 1 1 — — — 2 —
Key: An, anoestrus; Di, dioestrus; O, oestrus; P, pregnant; Y, yes; N, no; US, ultrasound; Lap, laparoscopy; CH no., corpus haemorrhagicum number; CL no., corpus luteum number; ND, not done due to advanced pregnancy.
R.M. Kirberger et al. / Theriogenology 76 (2011) 810 – 818
813
terone RIA kits (Progesterone Coat-a-Count, Diagnostic Products) as previously described [11]. 2.4.2. Vaginal cytology Vaginal smears were collected from each of the 12 sexually mature animals immediately prior to ultrasonography. These were obtained after passage of a sterilized, clear pyrex tube speculum of 12 mm diameter via the vulva and vestibulum. Thereafter, a sterile cotton-tipped swab attached to a stainless steel wire was passed through the tube until making contact with the cranial vaginal mucous membrane. The swab was rotated several times before retrieval and then rolled gently on a glass microscope slide, air-dried, and ® stained with Diff-Quik (Kyroquick, Kyron Laboratories, Benrose, South® Africa) before permanent mounting using Entellan (Merck, Halfway House, South Africa) and a glass cover slip. The slide was examined microscopically to describe the presence and percentage of superficial epithelial cells expressed as the Superficial Cell Index (SCI), and the clarity of the background as an indicator of oestrogenisation [1,2]. 2.4.3. Ovarian structures The ovaries of adult animals were observed by a combination of ultrasonography and laparoscopy in situ. The ultrasonographic variables recorded were ovarian dimensions (length, width, and height) and a prolate ellipsoid ovarian volume calculated using the method of Pavlik et al [12] for each observed ovary. All observed structures ⬎ 5.0 mm in diameter (follicles, CH, and CL) were recorded (Table 1). The ovaries were imaged on laparoscopic examination in all animals. Ovarian activity was defined by the observation of prominent structures. The presence of clear, vesicular appearing structures ⬎ 0.50 mm was recorded as follicles. The luteal phase was defined by observation of one or more prominent, somewhat irregular yellowish structures ⬎ 0.50 mm [13]. Based on the combination of observed ultrasonographic and laparoscopic ovarian structures, vaginal cytology, and the SPC, the non-pregnant lionesses were classified as being in oestrus (pro-oestrus), dioestrus, or anoestrus [7]. 3. Results 3.1. Ultrasound In the nine non-pregnant adult lionesses, six animals (66.7%) had visible uterine bodies dorsal to the bladder (Table 1). Of these, four (44.4%) had wall layering visible with the mean ventral and dorsal wall thickness being 3.3
Fig. 1. Uterine body ultrasonographic images (Lion 10): (A) sagittal image with near wall 4 mm and far wall 3.8 mm as indicated by cursors. Anechoic bladder in near field and hyperechoic line of ventral colonic wall just beyond uterus; (B) transverse image with total diameter of 10 mm between cursors. Bladder in near field and curved hyperechoic line of colonic wall ventrally and to right of uterus.
and 3.7 mm respectively (Fig. 1). Two of the uteri had anechoic content up to 1.7 mm wide and only one had a central echogenic line indicative of the wall-lumen interface but with no content present. Total mean uterine wall diameter was 7.9 mm (range 5.8 –10) and mean visible length was 43.9 mm. A separate vagina, cervix and uterine body could not be distinguished in any lioness. Terminal colonic gas interfered with visualizing the uterine body in one lioness. In only one lioness could both uterine horns be traced cranially, and then only for a short distance. These were both 7 mm wide and had a mixed echogenicity. In one lioness, both uterine horns and in another lioness one uterine horn could be traced caudally from the ovaries for up to 48 mm and in only one of these was wall layering distinct due to a small amount of anechoic seg-
814
R.M. Kirberger et al. / Theriogenology 76 (2011) 810 – 818
Table 2 Dimensions of 10* ovaries seen on ultrasound.
Length (mm) Width (mm) Height (mm) Volume (cm3)
Median
Min
Max
27.47 17.8 17.87 4.6
19 9.5 11.2 1.1
40 28.3 30.6 18.1
* Eleven ovaries were seen but only 10 measured.
mented fluid in the horn. This horn measured 10.1 mm in diameter whereas the other two horns were 6 and 6.6 mm in diameter, respectively. In the cubs, a uterine body could only be found in one cub where its diameter was 4.2 mm and it was visible for 30 mm dorsal to the bladder. No uterine horns could be seen. In the 12 adult lionesses, only 11 ovaries were located
and described of which in three lionesses both ovaries could be seen (Table 1). The observed ovarian dimensions are shown in Table 2. Eight ovaries had an irregular outline, each of which had at least a single obvious structure, i.e. follicle or luteal structure (CL or CH) protruding from the surface (Figs. 2–4). Two ovaries had a smooth outline of which one had a central 7.5 mm CH or CL whereas the other ovary had no obvious ultrasonographic evidence of cyclical activity. Follicles were seen on three ovaries, one of which was a very large structure with a diameter of 27.9 mm (Fig. 3), Structures classified as CH’s were seen on four ovaries and CL’s on two ovaries. The smallest observed structure was 6 mm in diameter. On the left side, colonic gas and on the right side, small intestines, respectively occasionally interfered with attempts to image the ovaries.
Fig. 2. Ultrasonographic (A and C) and laparoscopic images (B and D) (Lion 4): (A) left ovary 17.5 mm long diagnosed as having 1 follicle (F) caudally; (B) Left ovary diagnosed as having 2 follicles (F) located more caudally; (C) right ovary diagnosed as having 1 CH and several smaller unrecognized structures. Note the longitudinal image of a segment of small intestine in the near field (D) right ovary diagnosed as having 5 CH (3 larger ones annotated). Note the small periuterine cyst (arrow).
R.M. Kirberger et al. / Theriogenology 76 (2011) 810 – 818
815
and ovaries could still be seen while in several of the thinner lions some structures were still not visible. In cubs, the OI was much higher (median 0.34; range 0.22– 0.71 and no reproductive structures could be seen. 3.2. Laparoscopy
Fig. 3. Ultrasonographic and laparoscopic images (Lion 8): (A) right ovary (30.6 ⫻ 28.3 ⫻ 16.8 mm) diagnosed as having 1 large (27.9 mm diameter) and 2 small follicles (not in this image); (B) right ovary diagnosed as having 1 large follicle cranially.
In adults, the ovary was often located in the region where the caudal hyperechoic edge of the caudal perirenal fat joined the ventral hyperechoic surface edge of the iliopsoas muscle at an angle of about 60 degrees (Fig. 5). In adults the left ovary was on average 44.6 mm directly caudal to the left kidney and 50 mm caudal to the right kidney. In the cubs no ovaries were seen and the anatomical landmarks described above were poorly defined. Only three bladders were moderately filled, with the remainder having less urine volumes and mean bladder filling was 1.4. Data showed that the extent of bladder filling did not seem to influence visibility of the uterine body. The OI in adults ranged from 0.21 to 0.42 (median 0.27) and did not appear to influence the visibility of structures. In some obese lionesses the uterine body
Seven lions had a total of 12 active ovaries containing luteal structures in 10 ovaries and follicles in two ovaries (Table 1). One lion had two follicles on the left ovary and five CLs on the right ovary with a SPC of 94.5 nmol/l and no apparent signs of oestrogenisation on vaginal cytology (Lion 4). Two lions had small single periuterine cysts (Figs. 2D and 4B) and one lion had a cluster of periovarian cysts (⬎ 10) adjacent to each ovary (Fig. 6). Comparing ultrasonographic and laparoscopic ovarian findings in 11 lions that had undergone both procedures (Table 1), there was no evidence of cyclic ovarian activity (presence of follicles or luteal structures) in nine ovaries of six lions and two of these ovaries were found on ultrasound. Five lions had visible signs of reproductive activity (follicles or luteal structures) in six ovaries on both ultrasonography and laparoscopy and these modalities agreed 100% on the identification of the type of activity. Six ovaries of four lions had reproductive activity observed on laparoscopy but not on ultrasound and in two ovaries of two lions ultrasound detected reproductive activity with no activity found on laparoscopy (Table 1). Using laparoscopy as the gold standard to detect follicles and luteal structures, ultrasound had a sensitivity of 66% and specificity of 83% adult lionesses. 3.3. Reproductive status The SPC, vaginal cytology findings, and reproductive status are given in Table 1. Reproductive status of the mature lionesses (n ⫽ 12) was determined from a combination of data derived from ultrasonographic and laparoscopic observed ovarian activity, SPC, and vaginal cytology. In the prepubertal cubs (n ⫽ 5) SPC was a mean of 3.12 (range 2.9 –3.4) giving baseline reproductive activity values. Three lionesses were pregnant. Of the nine nonpregnant animals, two (22.2%) were in oestrus (follicular phase), three (33.3%) in the luteal phase (dioestrus), and four (44.4%) in anoestrus (Table 1).
816
R.M. Kirberger et al. / Theriogenology 76 (2011) 810 – 818
Fig. 4. Ultrasonographic (A and C) and laparoscopic images (B and D) (Lion 9): (A) left ovary (21.9 ⫻ 19 ⫻ 20.7 mm) diagnosed as having a single CH; (B) Left ovary diagnosed as having 3 CH (2 visible on this image). Note the small periuterine cyst (arrow); (C) right ovary (24 ⫻ 15.5 ⫻ 15.2 mm) diagnosed as having a single CH; (D) Right ovary diagnosed as having 1 CH. Bowel grasper and LigaSure® hand piece used for salpyngectomy.
4. Discussion This study was enhanced by direct laparoscopic visualisation to provide both a sensitive and specific comparative standard for concurrent ultrasonographic observations. This was reflected in the decreased sensitivity and specificity to detect follicles and luteal structures ultrasonographically as compared to laparoscopy. A further advantage of laparoscopic examination of the felid ovary is that the latter does not have an enveloping ovarian bursa (as in the Canidae), and this, together with manipulation of the ovary and its associated structures and adnexa, enables complete visualization of the ovary. The normal ovary without the presence of any distinct associated structures was difficult to visualize ultrasonographically and was only seen on the left side, where there is less small intestines, in two
lions. Similarly, the normal, non-gravid and particularly anoestrus uterus was difficult to visualize via ultrasound as previously reported [7]. In our reproductive clinic small animals are routinely examined in the standing position to allow the uterus to fall to the dependent surface to enhance visibility. This is impractical in the lion under anaesthesia and dorsal recumbency may thus exacerbate the difficulty in visualising the reproductive tract. Ultrasonographic detection of ovulation and hence definition of oestrous cycle status may be controversial to establish with great certainty on ultrasound alone [7,9]. However the sensitivity of 66% to detect follicles and luteal structures ultrasonographically in mature adult lions does imply that there is a place for ultrasound in evaluating the reproductive tract. Ultrasonography identified follicles on three ovaries, with laparoscopy
R.M. Kirberger et al. / Theriogenology 76 (2011) 810 – 818
Fig. 5. Ultrasonographic image of left ovary indicating location landmarks. Cursors indicate caudal perirenal fat with the caudal pole of the kidney indicated by short arrows. The long arrows are in the iliopsoas muscle with the tips at the ventral border of the muscle.
showing follicles on only two ovaries. It is assumed that ovulation did not take place in the interval between the completion of the ultrasound examination and the laparoscopic visualization of the ovary, which took about 30 min. A distinct ultrasonographic differentiation of the luteal structures (CL versus CH) was observed. Including the pregnant lions, six luteal structures were seen on six ovaries of five lions, i.e., never more than a single luteal structure per ovary, whereas laparoscopy showed 22 luteal structures on 10 ovaries in seven lions, i.e., an average of 2.2 per ovary. The OI was developed to assess if obesity played a role in hampering visibility of the reproductive structures, which it did not appear to do in this trial. Fat results in increased attenuation of the ultrasound beam and loss of image quality. However, in this study the presence of follicles or luteal structures appeared to be the most important factor to enhance ovary visibility. Operator experience (in this case a board certified radiologist) and good quality ultrasound equipment also enhance the ability to detect ovaries. The laparoscopic observation of the entire ovarian surface and the readily-identifiable luteal structures allows accurate calculation of an ovulation rate. This ovulation rate is dependent on the number of follicles successfully undergoing maturation, ovulation, and luteinization, and excludes anovulatory follicles. This observation in conjunction with observed litter sizes provides an improved index of both reproductive efficiency and fertility in this polytocous species as the
817
maximum litter size is dependent on the ovulation rate. The result is a ratio between litter size and ovulation rate in a particular oestrous cycle, rather than a categorical birth rate that suffices in monotocous species [13]. The number of CL’s observed suggest that on average 3.6 ovulations per cycle occurred in this population of pregnant and non-pregnant captive female lions. This is similar to the reported litter sizes in free-ranging populations in Southern Africa, where average litter sizes reported are 3.02 and 3.1, respectively, and a maximum of six cubs [14]. This observation implies the ovulation, fertilization, and litter rates, respectively, may be similar in this species as a whole. In the non-pregnant lions in this study the ovulations were presumably induced, albeit without actual coitus having occurred as these females were never in direct physical contact with any male lions, but male lions were certainly within close proximity in nearby enclosures. This observation shows a relatively high ovulation rate occurred without coitus in this population. There may obviously be a discrepancy in free-living populations where only data describing litter rates are reported, however, early neonatal attrition rates may be a confounding factor here. The SPC values must be interpreted in the light of variable endocrine results in immobilized or anaesthetized wild felids due to potential effects of these procedures on hormonal secretion [6]. This is also compounded by the observation of unexplained “pul-
Fig. 6. Laparoscopic image of a right ovary undergoing electrocoagulation of the uterine artery for ovariectomy. Note cluster of periovarian cysts caudoventrally to the ovary.
818
R.M. Kirberger et al. / Theriogenology 76 (2011) 810 – 818
sations” in circulatory SPC in lions [6]. The observation of active luteal structures via SPC, ultrasonography, and laparoscopy in three females without direct physical access to any adult male supports previous reports of spontaneous ovulation as a feature of the oestrous cycle in lions [1,3,4]. Laparoscopy showed that two lions had small single periuterine cysts and one lion had a cluster of periovarian cysts (⬎ 10) adjacent to each ovary. These were non-pregnant lions, giving a prevalence of 33% in non-pregnant lions in this study population. The significance of these structures is, however, uncertain and future laparoscopic studies may shed more light on their clinical relevance. In conclusion, transabdominal ultrasonography in combination with serum progesterone levels and vaginal cytology can be used to assess reproductive activity with some accuracy in captive bred lions. These procedures may offer potential benefits in assessment of individual reproductive function, including subfertility, and assisted reproduction techniques in similar populations of lions.
[2]
[3] [4] [5] [6]
[7] [8]
[9]
[10] [11]
Acknowledgments The authors thank E. Monnet, L. Venter, L. Bester, R. Serfontein, R. Fourie, and students who were involved with the ultrasonographic and laparoscopic procedures, as well as Karl Storz for sponsoring laparoscopic equipment. References [1] Schmidt AM, Nadal LA, Schmidt MJ, Beamer NB. Serum concentrations of estradiol and progesterone during normal oes-
[12]
[13]
[14]
trous cycle and early pregnancy in the lion (Panthero leo). J Reprod Ferti 1979;57:267–72. Bertschinger HJ, de Barros Vaz Guimaraes MA, Trigg TE, Human, A. The use of deslorelin implants for the long-term contraception of lionesses and tigers Wildlife Res 2008;35: 525–30. Brown JL. Female reproductive cycles of wild female felids. Anim Repr Sc 2010; doi:10.1016/Anireprosci.2010.08.024. Brown JL. Comparative endocrinology of domestic and nondomestic felids. Theriogenology 2006;6:25–36. Goodrowe KL, Feline reproduction and artificial breeding technologies. Jnl Repro Sci 1992;28:389 –97. Brown JL, Bush M, Packer C, Pusey AE, Montfort SL, O’Brien JL, Janssen DL, Wildt DE. Hormonal characteristics of free ranging female lions (Panthera leo) of the Serengeti plains and Ngorongoro crater. J Reprod Fert 1993;87:107–14. Davidson AP, Baker TW. Reproductive ultrasound of the bitch and queen. Top Comp An Med 2009;24:55– 63. Peter AT, Jakovljevic S. Real-time ultrasonography of the small animal reproductive organs. Comp Contin Educ Pract Vet 1992; 14:739 – 44. Root CR, Spaulding KA. Diagnostic imaging in companion animal theriogenology. Seminars in Vet Med and Surg (Small Animal) 1994;9:7–27. Smuts GL, Anderson JL, Austin JC. Age determination of the African lion 1978;185:115⫺46. Bertschinger HJ, Asa CS, Calle PP, Long JA, Bauman K, DeMatteo K, Jöchle W, Trigg TE, Human A. Control of reproduction and sex related behaviour in exotic wild carnivores with the GnRH analogue deslorelin: preliminary observations. J Reprod Fertil Suppl 2001;57:275– 83. Pavlik EJ, DePriest PD, Gallion HH, Ueland FR, Reedy MB, Kryscio RJ, van Nagell JR. Ovarian volume related to age. Gyn Oncol 2000;77:410 –2. Nothling JO, de Cramer KGM, Gerber D, Kammer VR. Luteal and follicular count in bitches: Assessment by means of magnetic resonance imaging. Theriogenology 2006;66:1343–54. Mills MGL, Bester MN. In: Mammals of the Southern African Subregion, Skinner JD and Chimimba CT (Eds) 3rd ed., Cambridge University Press, Cambridge; 2005, p. 396.