Long-term effect of neutering on plasma luteinising hormone concentration in cats

Abstract

Objectives

The objectives of this study were to validate a commercially available luteinising hormone (LH) cat ELISA, to determine whether the increases in plasma LH concentration that occur after neutering are maintained throughout cats’ lives and if other factors such as calendar seasons in both intact and neutered cats, and neutering age in neutered cats, influence plasma LH concentrations.

Methods

Stored plasma samples from client-owned cats were used for the measurement of LH concentrations. Clinical data, including age, sex, age at neutering and medical history, were reviewed. Two populations were included in this study: (1) a senior and geriatric cat population (⩾9 years old), including 18 intact and 18 neutered cats matched for age, sex and month of sample collection; and (2) an adult cat population (2–8 years old), including 45 neutered cats. LH concentrations were measured using a commercially available feline ELISA.

Results

Senior and geriatric neutered cats had higher plasma LH concentrations than age-matched intact cats (P <0.001). Calendar season did not influence plasma LH concentrations in the adult (P = 0.727) or senior/geriatric (P = 0.745) cats included in this study. No influence of age at neutering was observed on plasma LH concentrations (P = 0.296).

Conclusions and relevance

Neutering causes a significant long-term increase in LH concentrations in cats and further studies are required to determine the consequences on feline health.

Introduction

The practice of neutering cats has only become commonplace since the 1960s;^1–3 however, neutering is now the most common surgery performed in small animals.^4–6 A recent epidemiological study of client-owned cats from the USA reported that 81% of female cats and 83% of male cats were neutered. ⁶

The benefits and risks of neutering have been documented in both dogs and cats.^4,5,7 Societal benefits of elective gonadectomy include reduction of the number of relinquished animals by avoiding pet overpopulation.^1,8,9 Undesirable behaviours in male cats, such as aggression, roaming and urine spraying, can also be prevented with castration. ¹⁰ Furthermore, neutering has also been associated with a decreased incidence of certain diseases (eg, mammary, ovarian and uterine neoplasia, and pyometra). ¹ Conversely, an increased risk of the development of obesity and diabetes has been shown in neutered cats.^4,5 Additionally, the prevalence of certain types of cancer, including osteosarcoma, haemangiosarcoma, transitional cell carcinoma and prostatic neoplasia, have been shown to increase in neutered dogs, although whether neutering also increases cancer risk in cats is as yet unknown.^4,5

The short-term physiological effect of neutering in cats has been studied previously.^11–15 It has been shown that there is a rapid (<5 days) increase in luteinising hormone (LH) in response to ovariohysterectomy but that concentrations of this hormone, measured by ELISA, may fall again within 120 days.^11,12 However, studies using a semi-quantitative chromatographic assay (Witness LH test) have, in apparent contradiction, found that LH concentration remains above 1 ng/ml for many months, or even years, after neutering.^13–15

Chronic increases in circulating LH concentrations, along with other endocrine disturbances have been implicated in the pathogenesis of various diseases in postmenopausal women. For example, metabolic syndrome and cardiovascular disease have been linked to decreased plasma oestradiol and increased plasma LH concentrations.^16,17 In addition, LH receptors in the adrenal glands have been implicated in adrenocorticotropic hormone-independent hypercortisolism and hyperaldosteronism in postmenopausal women. ¹⁸ The first step in determining whether LH might play a role in disease pathogenesis in cats is to determine whether the increases in hormone concentration are maintained long term after neutering and whether they occur in both sexes.

Therefore, the objectives of the current study were as follows: (1) to validate a commercially available LH cat ELISA; (2) to determine the long-term effect of neutering on plasma LH concentrations in senior and geriatric cats; (3) to determine the influence of calendar season on plasma LH concentrations in intact and neutered cats throughout their lives; and (4) to determine the impact of neutering age on long-term LH concentrations. Our hypothesis was that LH concentrations would remain increased in neutered cats throughout their life, regardless of their sex, age at neutering or time of year during which the samples were collected.

Materials and methods

Validation of an LH cat ELISA

A validated commercially available assay marketed for measurement of LH concentrations in feline plasma (LH Cat ELISA kit; Abnova) was used in this study. This sandwich ELISA kit was composed of precoated plates with polyclonal anti-LH capture antibodies and used mouse monoclonal anti-LH secondary antibodies. In-house validation of the assay using the samples included in this study was performed. For each sample, plasma LH concentration was measured always in duplicate and therefore each measurement was itself a replicate. Coefficients of variation (CV) were calculated and measurements were considered adequate for statistical analysis if the CV was <15%. Precision and reproducibility of the assay was determined by calculation of intra- and inter-assay CV. Intra-assay variation was calculated by measuring the CV for samples measuring 0.5, 1.5 and 2 ng/ml by the same assay. Inter-assay variation was calculating by measuring the CV for samples measuring 0.5 and 3.7 ng/ml by different assays. The CV was calculated by dividing the standard deviation of the duplicate samples of the same concentration, by the mean and multiplying it by 100. This was repeated 2–4 times. Dilutional linearity was also determined by diluting a sample with the zero standard, obtaining three readings on the standard curve, which were compared with predicted concentrations. Mean recovery was calculated by dividing the observed concentration after dilution by the predicted concentration after dilution and multiplying it by 100. Potential cross-reactivity with other glycoprotein hormones was assessed using physiological and supraphysiological concentrations of thyroid stimulating hormone (TSH) of bovine pituitary origin (T8931; Sigma-Aldrich), human recombinant TSH beta (β) subunit (HOR-012; Prospec Protein Specialists) and human recombinant follicle-stimulating hormone (FSH)-β subunit (HOR-253; Prospec Protein Specialists). The limit of detection of the assay was 0.5 ng/ml, as reported by the manufacturer. For statistical analysis and graphical representation, any undetectable LH concentration was analysed and displayed at 0.25 ng/ml. In an attempt to improve the sensitivity of the assay, an extra standard (0.5 ng/ml) was added to those provided with the kit (0, 1, 2, 5, 10 and 25 ng/ml).

Data collection

Two study populations were included in this study: (1) a senior and geriatric cat population; and (2) an adult cat population.

Senior and geriatric cat population

A population of senior and geriatric cats was identified by retrospectively reviewing clinical records from a clinic based at two first-opinion practices in London, UK, between January 1998 and April 2018. This identified 18 intact cats, from which multiple stored heparinised plasma samples were available. In this paper the term ‘intact’ refers to non-gonadectomised cats. The clinic recruits and monitors healthy cats older than 9 years of age and cats with any of the following conditions, independent of their age: chronic kidney disease, hyperthyroidism and hypertension. Blood samples are routinely collected from cats attending the clinic for general health screening or monitoring of the previously mentioned conditions. Blood samples are obtained by jugular venepuncture, transported in a cooled bag and centrifuged prior to being submitted to a commercial laboratory for analysis. Residual blood samples are routinely stored at –80^°C at our institution and these were used for this study.

Clinical data such as age, breed, sex and sample collection date were obtained for these 18 cats, if available. Eighteen control samples from neutered cats matched for age, sex and month of sample collection were also identified from the same records and included in the study. Clinical data were collected as for the intact cat population. Of these 36 cats, 20 were female (10 intact and 10 neutered) and 16 were male (eight intact and eight neutered). Cats were of the following breeds: domestic shorthair (DSH; n = 27), domestic longhair (n = 4), Siamese (n = 2), Burmese (n = 1), Exotic Shorthair (n = 1) and British Shorthair (n = 1). Plasma samples were used for the measurement of LH concentrations. All cats from this population had more than one blood sample collected at different visits, which allowed for longitudinal LH concentration variation analysis. However, the length of time between visits was quite variable, ranging from 4 to 28 weeks. The number of samples provided by each group (intact and neutered cats) was the same. The effect of different calendar seasons on plasma LH concentration, including of the reproductive season, was also studied. The reproductive season for London, UK, was taken to be between 15 March and 20 August every year.

Adult cat population

A population of 45 neutered adult cats was also studied. All cats were between 2 and 8 years of age and were prospectively recruited by the same clinic for a different study investigating risk factors for development of hyperthyroidism. Twenty-four cats (53%) were male and 21 (47%) were female. Cats were of the following breeds: DSH (n = 23), Siamese (n = 9), Siamese cross (n = 1), Burmese (n = 9) and Tonkinese (n = 3). All cats were deemed healthy based on their medical history, routine physical examination (which also included measurement of non-invasive systolic blood pressure), serum biochemical analysis and urinalysis. Clinical data such as age, breed, sex, neutering date and date of sample collection were recorded when available. Archived frozen heparinised plasma samples were used for the measurement of LH concentration. Blood samples had been obtained and stored as for our senior/geriatric cat population; however, only one blood sample was available per cat, so no longitudinal analysis was possible. This second population was included with the aim of determining whether calendar seasons had an impact on LH concentration patterns in a younger population, as well as to investigate whether age at neutering affects plasma LH concentration. Information regarding neutering date was available for 28 cats.

Ethical approval and informed consent

The Ethics and Welfare Committee at our institution approved data collection, blood sampling and storage from all cats of the senior/geriatric population included in this study (Ethics Committee Approval number 20131258) and informed consent was obtained from respective owners. The adult cat population data and blood samples were obtained and stored under the Animals in Scientific Procedures Act with ethical approval from our institution (Ethics Committee Approval no. 20151415) and, again, informed consent was obtained from owners.

Statistical analysis

Statistical analysis and graphs were performed using statistical software packages (SPSS Statistics version 24 and GraphPad Prism 7). Statistical significance was determined as P <0.05. The distribution of numerical variables was assessed for normality by Shapiro–Wilk test and visual inspection of histograms. LH concentrations were compared between intact and neutered cats, male and female cats, and between visits (variation analysis) using the Mann–Whitney U-test. For the variation analysis the maximum difference between visits was determined for each individual cat. Individual differences in plasma LH concentrations were compared between intact and neutered cats. A mixed model was used to compare differences between calendar seasons (spring, summer, autumn and winter). The Mann–Whitney U-test was also used to compare LH measurements from samples taken during and outside the estimated feline reproductive season in London, UK. LH concentrations were compared between cats neutered before 4 months of age, between 4 and 6 months of age and after 6 months of age, using the Kruskal–Wallis test. Results are reported as median (interquartile range [IQR]), unless stated otherwise.

Results

In-house LH assay validation

Intra-assay CVs for samples measuring 0.5 ng/ml, 1.5 ng/ml and 2 ng/ml were 14.2%, 12.3% and 6.2%, respectively (n = 2–4). Four samples with undetectable LH concentration (0.25 ng/ml) were re-measured using the same and a different kit, and the same result was obtained. Inter-assay CVs for samples measuring 0.5 ng/ml and 3.7 ng/ml were 26.7% and 7.4%, respectively (n = 2–4). Dilutional linearity was assessed by performing two consecutive 1:2 dilutions (1:2 and 1:4 of the original sample) of a plasma sample with a LH concentration of 4.49 ng/ml, revealing mean recovery rates of 77.5% and 87.4%, respectively. The addition of TSH of bovine pituitary origin to the ELISA at supraphysiological concentrations of 0.5 ng/ml and 1 ng/ml resulted in measured LH concentrations of 10.97 ng/ml and 20.45 ng/ml, respectively. However, when using human recombinant TSH β-subunit at supraphysiological concentrations of 0.125 ng/ml, 0.25 ng/ml, 0.5 ng/ml and 1 ng/ml, no cross-reactivity occurred. The same was observed with human recombinant FSH β-subunit at physiological and supraphysiological concentrations of 1.5 ng/ml, 3 ng/ml, 6 ng/ml, 12 ng/ml and 24 ng/ml, which did not yield detectable concentrations of LH in this assay.

Senior and geriatric cat population

The distribution of plasma LH concentration was found to be non-Gaussian (Shapiro–Wilk test P <0.001), with only 34/106 measurements being above the limit of the detection of the assay (0.25 ng/ml). Plasma LH concentration was compared between intact and neutered senior/geriatric cats of both sexes. Senior and geriatric neutered cats had higher plasma LH concentrations (0.25 ng/ml [IQR 0.25–2.1 ng/ml], 25/53 measurements were detectable) than age-matched intact cats (0.25 ng/ml [IQR 0.25–0.25], 9/53 measurements were detectable [P <0.001]; Figure 1). LH concentrations in neutered cats were more variable than in intact cats (P = 0.0018; Figure 2). Twenty-eight of 53 measurements of neutered cats were undetectable, whereas 44/53 measurements of intact cats were undetectable. At visit 1, 14/18 intact cats had undetectable plasma LH concentrations, whereas only 6/18 neutered cats had undetectable measurements. At visit 2, 13/18 intact cats had undetectable plasma LH concentrations, whereas 11/18 neutered cats had undetectable measurements. At visit 3, all 17 intact cats had undetectable plasma LH concentrations, whereas 12/17 neutered cats had undetectable plasma LH concentration measurements. No difference was found in plasma LH concentrations across the different calendar seasons of the year (P = 0.745; Figure 3a) and plasma LH concentrations during the estimated feline reproductive season in London (0.25 ng/ml; IQR 0.25–0.49) did not differ from the measurements taken outside of this season (0.25 ng/ml; IQR 0.25–0.65 [P = 0.591]; Figure 4a). Also, no difference was found between plasma LH concentration between males (0.25 ng/ml; IQR 0.25–0.29) and females (0.25 ng/ml; IQR 0.25–0.68 [P = 0.123]).

Figure 1

Plasma luteinising hormone (LH) concentration in intact and neutered senior/geriatric cats (n = 36). The horizontal lines represent the median (dashed black line) and 25th and 75th percentiles (continuous coloured lines). Plasma LH concentration in neutered cats was significantly higher than in intact cats (P <0.001)

Figure 2

Variation of plasma luteinising hormone (LH) concentration between visits 1, 2 and 3 in neutered (orange) and intact senior/geriatric (purple) cats. Straight lines connect plasma LH measurements of the same cat. Plasma LH concentrations in neutered cats fluctuate significantly more than in intact cats (P = 0.018)

Figure 3

Plasma luteinising hormone (LH) concentration of the (a) senior/geriatric cat population and (b) adult cat population across the different seasons of the year. Note the different scale of the y axes. The horizontal lines represent the median (dashed black line) and 25th and 75th percentiles (continuous coloured lines). Plasma LH concentration was not different across the different calendar seasons in senior/geriatric (P = 0.745) or adult (P = 0.727) cats

Figure 4

Plasma luteinising hormone (LH) concentrations across the different months of the year of the (a) senior/geriatric cat population and (b) adult cat population. The area within the orange rectangle refers to the estimated feline reproductive season in London, UK. ¹⁹ No difference was found between plasma LH measurements during and outside of the reproductive season in senior/geriatric (P = 0.591) and adult (P = 0.474) cats. Neutered cats are represented as orange dots and intact cats as purple dots. Some measurements of LH are superimposed, particularly for lower concentrations

Adult cat population

Plasma LH concentration was again found to be non-normally distributed (Shapiro–Wilk test P <0.001), with most cats having undetectable plasma LH concentration (n = 36/45). As in the senior/geriatric cat population, no difference was found in plasma LH concentrations across the different calendar seasons of the year (P = 0.727; Figure 3b). Again, plasma LH concentrations during estimated local reproductive season (0.25 ng/ml [IQR 0.25–0.29 ng/ml]; 18/24 cats had undetectable LH concentration) did not differ from the measurements taken outside of this season (0.25 ng/ml [IQR 0.25–0.25 ng/ml]; 18/21 cats had undetectable LH concentration [P = 0.474]; Figure 4b). Similarly, no difference was found between LH concentrations between males (0.25 ng/ml; IQR 0.25–0.25 ng/ml) and females (0.25 ng/ml; IQR 0.25–0.282 [P = 0.696]) of the adult cat population. No influence of age at neutering was observed on plasma LH concentration (P = 0.296; Figure 5), when the latter was compared between cats neutered before 4 months, between 4 and 6 months and after 6 months of age.

Figure 5

Plasma luteinising hormone (LH) concentration of adult cats (n = 28) that had been neutered before 4 months, between 4 and 6 months and after 6 months of age. The horizontal lines represent the median (dashed black line) and 25th and 75th percentiles (continuous coloured lines). Age at neutering did not affect long-term plasma LH concentration (P = 0.296)

Discussion

To our knowledge, this is the first study to describe plasma LH concentrations in intact and neutered senior/geriatric cats. As hypothesised, neutering has a long-term effect on circulating LH concentrations in cats, with neutered senior and geriatric cats having significantly higher concentrations than age-matched intact cats. Our results appear to refute the previous suggestion that the increase in LH following gonadectomy in cats is only transient as raised plasma LH concentrations were detected in adult, senior and geriatric neutered cats. ¹¹ Although direct comparisons were not performed between cats of the senior/geriatric population and adult cat population, older neutered cats seem to be more likely to have detectable LH concentrations than adult neutered cats. However, this may be the result of LH having been measured multiple times in cats from the senior/geriatric population and only once in cats from the adult population.

Assay validation steps revealed acceptable precision and dilutional linearity with a mean recovery of >80%. However, some sample matrix effect may have contributed for it not to be higher. Investigations performed with regard to possible cross-reactivity between glycoprotein hormone detection by this ELISA kit revealed that this is minimal when the beta subunit of recombinant human TSH and FSH is used. It is suspected that the apparent cross-reactivity when using TSH from bovine pituitary may represent contamination with LH from the pituitary gland. Furthermore, the TSH concentrations used were substantially higher than the median TSH concentration of euthyroid cats (0.05 ng/ml). ²⁰ Various studies have shown that repeated freezing and thawing does not affect the measured concentration of several commonly studied reproductive hormones, including LH, and therefore this was not tested as part of the validation process.^21–23

LH is secreted in a pulsatile manner and is believed to be stimulated by many environmental factors such as photoperiod/season, stress, immune and nutritional status.^24,25 In our study, we documented considerable variation of plasma LH concentrations in neutered cats, even to low or undetectable concentrations similar to those found in intact cats, which may, at least in part, reflect its pulsatile release. To date, studies characterising plasma LH concentration variation in cats have mainly focused on the timing of the LH surge in relation to copulation and ovulation and very little is known about the variation of the concentrations of this hormone outside of the oestrus period or in cats that have been neutered.^11,26,27 Although not exhaustibly, our study provides some insight into the variation of plasma LH concentrations in client-owned cats living in an urban area (London).

Our findings indicate that a single measurement of plasma LH concentration as a method to determine if a cat is neutered may be unreliable. Neutered cats with transiently low LH concentrations may be interpreted as being intact if they have serum LH concentrations below the cut-off of semiquantitative LH assays clinically available (1 ng/ml) and used in previous publications. Therefore, it would be advisable to repeat the assay a few days later and/or to use another method for determining the serum/plasma concentration of LH and/or to investigate for the presence of a surgical scar or absence of reproductive organs on abdominal ultrasound examination suggestive of previous gonadectomy.^13–15 Similarly, queens in oestrus may be considered to be neutered if they have serum LH concentrations >1 ng/ml. It would therefore be advisable to perform vaginal cytology at the time of LH measurement in order to identify queens that may be in oestrus. ¹⁵ An important point related to the semiquantitative assays used to determine if a cat has been neutered is that any plasma LH concentrations <1 ng/ml are given as ‘not being neutered’. However, in our study we found that neutered cats frequently have undetectable plasma LH concentrations or concentrations of LH <1 ng/ml. We had intended to run plasma LH semiquantitative assays, as described in the referenced literature, with our samples with the aim of obtaining information on how it compared with the ELISA used in our study, especially given that the limit of detection of the latter was much lower (0.5 ng/ml). However, the semiquantitative assay was not available at the time that this study was conducted.

In this study we had proportionally fewer cats with plasma LH concentration >1 ng/ml when compared with the results of the publications previously discussed.^13–15 This could be related to the fact that the assay used in this study may not be adequately optimised for use in cats and that a different assay might have yielded more samples >1 ng/ml. However, it is important to note that the described studies using the semiquantitative LH assay (Witness LH test), were mainly performed in young female cats, susceptible to LH surges if intact.^13–15 In addition, it is possible that the Witness LH test could cross-react with other glycopeptide hormones.

Our study found no influence of calendar seasons on plasma LH concentrations in either adult, senior or geriatric cats, with no difference being observed across the different seasons of the year or during the estimated feline reproductive season. The reproductive season for London, UK, was determined using the diagram published by Hurni, which describes the duration of the reproductive season for cats living under natural conditions in relation to geographical latitude. ¹⁹ For London, the reproductive season was estimated to be between 15 March and 20 August every year. It is important to note that all intact cats included in this study were senior or geriatric and some of them suffering from concomitant diseases such as hyperthyroidism, chronic kidney disease and hypertension. Consequently, this may have influenced their response to photoperiod and ability to produce an LH surge.

One of the limitations of this study was the lack of available information regarding the oestrous cycle phase of the female cats included, which would have been interesting to put into perspective with plasma LH concentration measurements. LH concentration in cats during a surge has previously been reported to reach 200–300 ng/ml, much higher than those documented in any cat included in our study. ²⁷ Interestingly, a study performed in male dogs documented exactly that. Senior intact male (11 years old) dogs have a reduced number of LH surges (four per day), when compared with adult male dogs (4–7 years old), which had 7–8 LH surges in 24 h. ²⁸ The lack of an effect of photoperiod in our study is perhaps surprising as this has been shown to determine the beginning and end of the breeding season in cats, with cats in the tropics breeding year-round and those residing towards the poles having only 1–2 litters per year.^26,29–33 However, the majority of cats included in our study (n = 63/81) were neutered and so the power to detect changes with photoperiod in older intact cats was limited. Furthermore, it is also possible that should we have analysed more samples from each cat, over different calendar seasons, we could have found an effect of photoperiod on plasma LH concentrations.

The impact of neutering age on long-term LH concentrations was also investigated. Cats usually reach puberty between 5–9 months of age, when their body weight is 75–80% of their adult body weight (2.3–3.2 kg). ²⁶ However, this is also dependent on the breed, time of the year and photoperiod, social environment, health and nutritional status.^26,30 We hypothesised that early neutering could influence cats’ ability to produce LH post-gonadectomy owing to possible lack of maturation of the hypothalamic–hypophyseal–gonadal axis. Interestingly, we found no difference in plasma LH concentrations between cats neutered before 4 months, between 4 and 6 months and cats neutered after 6 months of age. However, it is important to note that the majority (n = 25/28) of the cats included in this study were neutered after 4 months of age, so this study was not well-powered to evaluate the effect of early or pre-pubertal neutering.

Conclusions

Neutering causes a significant long-term increase in plasma LH concentrations in cats and, therefore, further research to determine its consequences to feline health is warranted. Although higher concentrations of plasma LH concentrations are found in senior/geriatric neutered cats than in intact cats of the same age, this increase is not constant. Given the pulsatile secretion of LH, plasma LH concentration in neutered cats can be as low as the concentrations found in intact cats and so should not be used in individual patients as a reliable indication of neuter status.