Prevalence, morphology and associated risk factors for cystic ovaries in companion guinea pigs (Cavia porcellus)

Abstract

Background

Ovarian cysts are the most common reproductive pathology in guinea pigs, yet their prevalence and associated risk factors in clinically healthy individuals remain poorly understood. This study investigated the prevalence, size and morphology of ultrasonographically detectable ovarian cysts in guinea pigs and their associated risk factors.

Methods

A total of 221 female guinea pigs from private households and breeding collections underwent ultrasound examination to identify ovarian cysts and determine their size and morphology. Each animal's age, breed, parity and body condition were also recorded.

Results

The overall prevalence of ovarian cysts was 41.6%, increasing from 25.4% in animals less than 1 year old to 68.2% in those over 3 years old. Cyst size increased significantly with age (0.33 mm/month). Each increase in parity was associated with a 2.15 mm decrease in cyst size. Guinea pigs from breeders had a significantly lower prevalence of polycystic ovaries compared to privately owned individuals. Breed and body condition did not significantly affect cyst prevalence.

Limitations

Histopathological cyst type and concurrent uterine disease were not examined.

Conclusion

Age was the primary risk factor for ovarian cyst development in this population of guinea pigs, while an increase in parity decreased cyst size. Further research is needed to explore additional risk factors and establish clinically relevant cyst size thresholds.

INTRODUCTION

The guinea pig ovary often serves as a comparative model for human ovarian pathology; and while factors such as body condition, endocrine imbalances, and possibly genetics influence the prevalence of cystic and polycystic ovaries in women, this remains to be investigated in guinea pigs. ¹ , ² Intraovarian cysts are the most common pathology of the female reproductive tract in guinea pigs, with reported prevalence ranging from 8.6% to 100% depending on the focus of the study, animal age and method of cyst detection. ³ , ⁴ , ⁵ , ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ Reported cyst sizes in guinea pigs range from 0.5 to 80 mm, and while larger cysts can be detected through abdominal palpation, ultrasonographic evaluation of the ovaries remains the gold standard for non‐invasive diagnosis in the live animal. ³ , ¹⁶ , ¹⁷ The sensitivity and specificity of cyst detection depend on the examiner and imaging depth, with a higher sensitivity and specificity being reported for cysts measuring 4 mm or more. ¹⁸ However, even smaller cysts can be detected with histopathology, increasing sensitivity and resulting in a prevalence of up to 100%. ³ , ¹⁴ Histopathology in combination with immunohistochemistry also allows for differentiation of cyst types, making histopathology the preferred diagnostic tool for diagnosing and characterising ovarian cysts in the surgically excised ovary or at necropsy. ³ , ⁴ , ⁹ , ¹¹ , ¹⁵ , ¹⁹ Most cysts are rete cysts—a possibly normal physiological component of the cycling ovary. Follicular cysts, on the other hand, are functional cysts of unknown pathogenesis and prevalence, but are less frequently reported. ¹⁰ , ²⁰ , ²¹ , ²² , ²³ , ²⁴ , ²⁵ Although rarely differentiated from single cysts, polycystic ovaries have been reported in up to 96.4% of cases, all being rete cysts. ²⁶

As a prey species, clinical signs of discomfort in guinea pigs are subtle and often go unnoticed by the owner. Clinical signs related to ovarian cyst size are usually absent until the cyst has grown considerably, causing discomfort by compressing surrounding anatomical structures, or until cyst rupture causes cystic fluid to leak, leading to local peritoneal inflammation. ²⁷ , ²⁸ Hormone‐producing follicular cysts may cause more overt clinical signs, such as hyperandrogenism, manifesting as alopecia, clitoral hypertrophy, aggression and mammary hyperkeratosis. ¹³ While rete cysts were originally believed to be non‐functional, oestradiol and progesterone have recently been confirmed in cyst fluid, challenging this theory. ²⁹ Clinical signs can also arise from concurrent reproductive tract comorbidities such as endometrial hyperplasia, mucometra, neoplasia or endometritis, which are reported in up to 97.1% of affected guinea pigs. ³ , ³⁰ , ³¹ While percutaneous drainage of large cysts may cause temporary relief, this procedure is not curative and carries significant risks. ²⁷ Hormonal treatments show variable efficacy but may only affect the presence of follicular cysts. ²⁷ , ³² , ³³ , ³⁴ , ³⁵ Due to frequent concurrent uterine pathology, treatment by ovariohysterectomy remains the gold standard. ²⁷ , ³⁶ However, because large ovarian cysts are often detected in older animals with comorbidities and reproductive tract surgery is complicated in female guinea pigs, surgical risks are considerable. ⁸ , ⁹ , ²⁶ Therefore, identifying risk factors is critical in the search for preventive measures and to enable earlier, less invasive treatment options.

The objective of this study was to investigate the prevalence, morphology and size of ovarian cysts detectable by ultrasound in intact female guinea pigs of different ages and breeds, which are representative of the pet guinea pig population presented in clinical practice. The influence of age, breed, parity, body condition score (BCS) and origin on cyst prevalence was also assessed.

MATERIALS AND METHODS

Intact, non‐pregnant female pet guinea pigs (Cavia porcellus) presenting to the Exotics Department at AniCura Copenhagen Animal Hospital between March 2017 and February 2021 were included in the study. These animals were either patients or companions to hospitalised guinea pig patients. Animals of all ages and parities were included. Guinea pigs previously diagnosed with ovarian disease, treated with hormones, or referred for treatment of cystic ovaries were excluded, to not overestimate the disease prevalence in the population. Additional guinea pigs were recruited via social media from Danish guinea pig breeders registered with the Danish Cavy Club. In breeding collections with fewer than 10 females, all individuals were included to avoid breeders positively selecting guinea pigs in their collection with suspected ovarian cysts. In larger breeding collections, a representative convenience sample across age, breed and parity was selected, based on how many guinea pigs the breeder could transport to the hospital. Pregnant, lactating or unweaned animals were excluded to minimise stress. Age (in months), parity and presence of non‐pruritic alopecia of the flanks were noted for all the animals. BCS (1‒5) was noted in a subgroup of animals at the end of the study period and was grouped for statistical analysis as underweight (BCS 1 and 2), ideal weight (BCS 3) or overweight (BCS 4 and 5). ³⁷ Breed was recorded in the last half of the project based on phenotypic appearance according to the Danish Cavy Club guidelines. Breeds (as recognised by the Danish Cavy Club) were grouped for statistical analyses, based on fur texture, length and swirl characteristics. Shorthaired breeds were grouped as teddy, Abyssinian or American regardless of colour. Peruvian, coronet and texels were grouped as longhaired animals. Animals with a satin phenotype were grouped as silkies, regardless of colour or fur length. Information on the animals’ age, breed, parity, origin and BCS was unknown to the ultrasonographer.

The ovaries were evaluated ultrasonographically for the presence of cysts using a flank approach. The non‐sedated guinea pigs were placed on a non‐slippery surface with all feet on the table, before parting the fur of the flanks manually and applying alcohol and ultrasound gel. Excellent image quality could be obtained with this approach. The stress response usually observed when placing guinea pigs in dorsal recumbency for ultrasonography was not observed when using the flank approach. The landmark for the position of each ovary caudal to the kidney was examined for cysts with ultrasound (GE Logiq E9, GE Healthcare), using a microconvex probe at 10 MHz (GE C3‐10, GE Healthcare). Anechoic circular structures in relation to the ovary were interpreted as ovarian cysts (Figure 1). ¹⁷ When present, the maximum diameter of the ovarian cyst was recorded in millimetres. ¹⁸ If multiple cysts were detected on one ovary, the maximum diameter of each individual cyst was noted and summed, and the ovary was reported as polycystic. The total cyst size per animal was calculated as the sum size of all cysts on both ovaries.

FIGURE 1.

Examples of ultrasonographic images of ovarian cysts (a and b) and polycystic ovaries (c and d) in female guinea pigs. One cyst is present on the right ovary in (a and b), while two cysts are present (polycystic) on the left ovary in (c) and right ovary in (d).

Data analysis

Statistical analyses were computed in R (version 4.0.3). ³⁸ Generalised linear models (GLMs) were used to perform binomial logistic regression (logit model) and analyse the influence of a series of predictor variables—age, parity, breed and origin (breeder vs. private household)—on the presence of ovarian cysts. An interaction between breed and origin was included to account for selective breeding practices. To evaluate if BCS influenced the presence of cysts, a separate GLM was performed on a subset of the data for which measures of BCS existed (65 animals). This GLM had the same structure as above, except that BCS was included as an additional predictor variable. Prevalence ratios (PRs) and 95% confidence intervals (CIs) were estimated from these logistic models (package prLogistic). ³⁹ For animal with cysts, a linear model (LM) was used to analyse the effect of age and parity (predictor variables) on cyst size (response variable). Model reduction was done by stepwise removal of non‐significant predictors (p > 0.05), with likelihood ratio tests (LRTs) used for comparison (package lmtest). ⁴⁰ For categorical variables with more than two levels, LRTs compared full and reduced models to assess the overall effect. The results related to age are reported as median and interquartile range (IQR), while those related to cyst sizes are reported as mean ± standard deviation (SD).

RESULTS

Two hundred and twenty‐one intact female guinea pigs ranging in age from 1 to 82 months (median = 20 months; IQR = 12‒33 months) were included. Of the 221 animals, 140 were recruited from nine Danish breeders. A total of 145 were nulliparous, while the remaining 76 had delivered a total of 137 litters. The number of litters per animal was divided as one (n = 35), two (n = 27), three (n = 11), four (n = 2) or seven (n = 1) litters. Breed was noted in 160 animals. The breeds represented were American (n = 100; 45.2%), teddy (n = 23, 10.4%), longhaired (n = 20; 9%), Abyssinian (n = 12; 5.4%) and silkie (n = 5; 2.3%). Breed was not noted in the remaining animals (n = 61; 27.6%).

Ovarian cysts were detected on 126 ovaries in 92 of the 221 animals (41.6%), ranging from 4 to 81 months of age (median = 25 months; IQR = 15‒42 months) (Figure 2). Unilateral cysts were found on the right (n = 26) and left (n = 32) ovary, while 34 animals had bilateral cysts (Table 1). Increasing age significantly increased the likelihood of cyst presence (GLM: z = 4.063, p < 0.001; PR = 1.029, 95% CI = 1.006‒1.046). In animals above 36 months of age, the prevalence of cystic ovaries increased markedly to 68.2% (Figure 2a). The overall occurrence of cysts was not significantly affected by origin (GLM: z = ‒0.974, p = 0.330) or breed (LRT: χ ² = 7.576, p = 0.181) (Figure 3).

FIGURE 2.

Prevalence of ovarian cysts (a) and polycystic ovaries (b) in 221 female guinea pigs of different ages. The summed prevalence with increasing age is shown in blue. Sample sizes (n) in each 1‐year bin are also listed. Cyst occurrence significantly increased with age (analysed as a continuous variable but presented here in 1‐year bins) for both cysts overall and polycystic ovaries specifically. GLM, generalised linear model.

TABLE 1.

Distribution and prevalence of cystic and polycystic on the right and left ovaries in 221 guinea pigs.

Prevalence of cystic changes in 221 animals	Right ovary
	No cyst	Single cyst	Polycystic
Left ovary
No cyst	129 (58%)	22 (10%)	4 (2%)
Single cyst	29 (13%)	19 (9%)	6 (3%)
Polycystic	3 (1%)	7 (3%)	2 (1%)

Note: Number of animals and the percentage (rounded) of the population are given.

FIGURE 3.

Prevalence of ovarian cysts in 221 female guinea pigs of different breeds. There was no significant effect of breed on the occurrence of cysts. The mean prevalence of cysts across all ages (41.6%) is shown in cyan (dashed line). Silhouettes of breed phenotype are shown above each column. LRT, likelihood ratio test.

Polycystic ovaries were observed in 22 animals aged 5‒81 months (median = 37 months; IQR = 29‒50 months). Bilateral polycystic ovaries were found in two animals, while 13 of 20 (59.1%) with a unilateral polycystic ovary had a single cyst on the contralateral ovary. Age also significantly increased the prevalence of polycystic ovaries (GLM: z = 3.538, p < 0.001; PR = 1.043, 95% CI = 1.015‒1.071; Figure 2b), while polycystic ovaries were less common in animals from breeders (GLM: z = ‒2.400, p = 0.02; PR = 0.311, 95% CI: 0.000‒0.639).

Single cyst sizes ranged from 2.2 to 37.2 mm on the right ovary (mean = 5.4 ± 6.4 mm) and from 3.2 to 37.3 mm on the left ovary (mean = 5.2 ± 5.6 mm). The summed diameter of cysts of polycystic ovaries was 4.5‒31.9 mm (mean = 12.5 ± 6.2 mm). Total cyst size significantly increased with age by 0.33 mm per month (LM: t = 5.669, p < 0.001; 95% CI = 0.22‒0.45) (Figure 4a), but significantly decreased with increasing parity by 2.15 mm per litter (LM: t = ‒2.470, p = 0.02; 95% CI = ‒3.87 to ‒0.42) (Figure 4b). No interaction effect of age and parity was found in relation to cyst size (LM: t = ‒0.258, p = 0.797). Cyst size in polycystic ovaries was not affected by age (LM: t = 0.670, p = 0.511) or parity (LM: t = ‒0.456, p = 0.653).

FIGURE 4.

Summed size of cysts on both ovaries for each female guinea pig with ovarian cysts (n = 92) increased significantly with age (A) and decreased significantly with parity (B) according to the linear model (LM): summed cyst size = 0.335 [0.059] × age − 2.146 [0.869] × parity + 3.411 [2.056] (values in square brackets are standard errors). Blue lines are predicted regression lines from the LM, surrounded by 95% confidence interval bands in light grey and predicted at the median value of the other predictor variable (i.e., the effect of age on summed cyst size is predicted at a parity of 0, and the effect of parity on summed cyst size at an age of 25 months). Cyan dots represent the summed cyst size for individual animals. The dots are semi‐transparent, so darker colour indicates a greater number of overlapping data points.

BCS was evaluated in 65 animals. None were underweight (BCS 1 or 2), 43 animals (66.2%) had an ideal BCS (BCS 3) and 22 animals (33.8%) were overweight (BCS 4 or 5). BCS was not associated with cyst presence (GLM: BCS with two levels, ideal or overweight; z = ‒0.080, p = 0.936). Only two animals with polycystic ovaries had BCS recorded. These were both overweight.

Bilateral non‐pruritic flank alopecia was observed in two animals (0.9%). These were 23 and 30 months old and had total cyst sizes of 3.9 mm (single cyst) and 16.4 mm (single + polycystic ovary), respectively. No blood analyses were performed to exclude other causes of flank alopecia.

DISCUSSION

This study found that 41.6% of female guinea pigs aged 1‒82 months had cystic ovaries, with size and prevalence significantly increasing with age. A smaller, but similar, study ultrasonographically screening a guinea pig population in 2003 found a higher cyst prevalence (58%) despite examining a younger breeding population (13 ± 16 months, reported mean ± SD) and having a higher cyst size inclusion threshold than in the present study (≥4 and 2.2 mm, respectively). ⁹ The lower overall prevalence of 41.6% in the present study may reflect both the exclusion of referred symptomatic cases and demographic population differences. No lateral predilection for cyst development was observed, in contrast to previous studies suggesting a right or left ovarian preference. ⁹ , ¹³ Bilateral cysts were present in one‐third of affected animals, similar to a previously reported ultrasound‐based screening study. ⁹

In the literature, the reported prevalence of ovarian cysts in female guinea pigs ranges from 8.6% to 100%. ³ , ⁴ , ⁵ , ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ This variation in cyst prevalence is likely caused by a number of factors, including inclusion criteria, age, method of cyst detection and the purpose of the study. Necropsy studies (biased toward older animals) report high cyst prevalence, with 75.6% of females over 6 years being affected, ¹⁵ while histopathology, being the most sensitive method, often reveals small cysts undetectable by ultrasound and confirms a near 100% prevalence. ⁷ , ¹¹ , ¹⁵ , ²³ Even though histopathology carries the highest sensitivity for the detection of cysts, ultrasonographic evaluation remains the gold standard for non‐invasive screening of the live animal. ²⁸

Polycystic ovaries are rarely distinguished from single cysts in the literature, although histological data suggest that they may be the most common (up to 96.4%). ²⁶ Polycystic ovaries, although increasing with age, were relatively uncommon in this study (10%). Among animals with unilateral polycystic ovaries, two‐thirds had a single cyst on the contralateral side, and only a minority of those had polycystic (9%). The lower prevalence of polycystic ovaries observed here may reflect limitations of ultrasonographic resolution in detecting thin septa within cystic structures. ⁴¹ However, the absence of histological confirmation limits this interpretation.

Overall, cyst prevalence was not significantly associated with breed or origin, suggesting that phenotype may not predict cyst formation. However, breed classification in this study was based on the fur type and texture rather than the colour pattern or genetic testing. Regrouping of breeds might reveal other inheritance patterns. While nearly all guinea pigs develop cysts histologically, ³ , ¹⁴ identifying potential inheritance patterns for follicular cysts (which are less common but clinically relevant) may be feasible in future studies, particularly through pedigree analyses. However, the practical implications are challenging in a breeding programme, since histological confirmation of cyst type is needed, making the animal infertile. Although origin (private household vs. breeder) did not influence overall cyst prevalence, polycystic ovaries were less common among guinea pigs from breeders. One possible explanation is unintentional deselection of animals with reduced fertility. As the summed cyst size of polycystic ovaries was larger, they may compromise ovarian function more significantly, resulting in infertility and exclusion from breeding programmes. ¹³ , ⁴² When excluded from the breeding programme, these animals are often adopted into private households, which could have increased the prevalence of animals with polycystic ovaries in this group.

The cyst size increased significantly with age, which is consistent with previous reports. ⁸ , ⁹ , ¹¹ , ¹³ , ¹⁵ Interestingly, cyst size decreased with increasing parity. This could reflect a selection bias, as animals with smaller cysts may be more fertile. Alternatively, the hormonal impact of pregnancy itself could influence cyst development. In women, being non‐nulliparous is associated with reduced risk of certain ovarian pathologies, a concept that warrants further investigation in guinea pigs. ⁴³ Among multiparous guinea pigs with cysts, 71.4% had only unilateral involvement, preserving one functional ovary. Given that a single ovary is sufficient for reproduction in women, it is possible that fertility in guinea pigs is similarly maintained when one ovary remains unaffected. ⁴⁴

Body condition did not significantly affect cyst presence or size in this study. This is contrary to prior findings. ¹¹ However, no underweight animals (BCS 1 or 2) were included in the present study, limiting the range of comparison. Underweight guinea pigs were rarely presented to the clinic, and when presented, these were most often moribund and not clinically stable for ultrasonographic evaluation. Additionally, cyst types were not histologically confirmed, so any relationship between BCS and follicular versus rete cysts could not be assessed.

Only two of the 221 guinea pigs (0.9%) in the study had bilateral, non‐pruritic flank alopecia, much lower than previous reports (4.7%). ⁹ This likely reflects study design, as animals referred for dermatological signs associated with ovarian cysts were excluded in this study. One guinea pig with alopecia came from a breeder and was included due to full‐collection screening, the other being a companion for an in‐patient. While the reported prevalence of dermatologic changes in guinea pigs with cystic ovaries is low, cystic ovaries are found in up to 6.2% of female guinea pigs with dermatological changes. ¹² Since rete cysts are very common in the population, and not always associated with dermatological changes, it is important to consider alternative endocrine disorders, such as hyperadrenocortisism and hyperthyroidism, in guinea pigs with both ovarian cysts and alopecia. ¹² , ²⁸ , ⁴⁵ , ⁴⁶

The limitations of the study include the likely underreporting of guinea pigs exhibiting dermatologic changes due to the inclusion criteria, which excluded patients referred for suspected ovarian cysts. Dermatological changes are among the most prominent clinical signs and are a common reason for referral to the authors’ clinic for suspected ovarian cysts. These animals were therefore excluded to prevent overestimating the prevalence of cysts in general and associated dermatological changes. While BCS was not found to be associated with cyst presence, the limited number of animals with recorded BCS, combined with the absence of underweight individuals, restricts the interpretability of these findings. Additionally, genetic factors may still play a role and warrant further investigation. Grouping breeds based on coat color patterns or pedigree could potentially reveal inheritance patterns related to cyst development and cyst types.

While increasing age was associated with a higher prevalence and size of both single and polycystic ovaries, increasing parity resulted in smaller cysts. Furthermore, polycystic ovaries were less common in guinea pigs from breeders. Assessing the clinical relevance of cysts remains challenging. In women, cysts larger than 4 cm in diameter are associated with pain. ⁴⁷ Relative to ovary size, this equates to ∼10.4 mm in guinea pigs. In this study, 24 animals (10.9%) had cysts at or above this threshold. Whether this corresponds to pain or discomfort in guinea pigs is unknown. Future research should aim to determine a clinically significant size threshold for intervention. This would help guide recommendations for ovariohysterectomy, especially in asymptomatic animals where pain may go unnoticed by owners.

AUTHOR CONTRIBUTIONS

Conceptualisation and data collection: Mette L. Halck and Frederik Meyland‐Smith. Data analysis: Tommy Norin and Mette L. Halck. Writing—original draft: Mette L. Halck. Writing—review: Frederik Meyland‐Smith and Tommy Norin.

CONFLICT OF INTEREST STATEMENT

The authors declare they have no conflicts of interest.

FUNDING INFORMATION

The authors received no specific funding for this work.

ETHICS STATEMENT

The study was approved by the Ethics and Administrative Committee, Department of Veterinary Clinical Sciences, Faculty of Health and Medical Sciences, University of Copenhagen, Denmark (#2021‐11). The publication of data in this study complies with the General Data Protection Regulation.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.