Quantifying the Role of Progestogen and Clucocorticoid Exposure in the Epidemiology of Feline Diabetes Mellitus in Romania
Background: Progestogen and glucocorticoid administration have been documented as a risk factor for the development of feline diabetes mellitus (DM) in a handful of studies. Despite this, they remain popular among veterinarians. Additional information regarding their exact diabetogenic risk could aid the risk-benefit assessment each clinician needs to make when considering their use.
Objective: To quantify the risk for feline DM associated with use of exogenous progestogens and glucocorticoids in Romania.
Methods: This retrospective cohort study used the clinical records of all cats presenting to the Faculty of Veterinary Medicine of Iasi, Romania, in the period of 2005-2014. The following parameters were assessed: signalment, weight, DM diagnosis and exogenous progestogens and/or glucocorticoids (including topicals) exposure. Patients were divided into five groups: PE: progestogen-exposed; GE: glucocorticoid-exposed; NPE: no progestogen-exposure; NGE: no glucocorticoid-exposure; NPGE: no progestogen or glucocorticoid-exposure. Period prevalence (PP) of DM in each group was calculated, as well as odds ratios (OR), including 95% confidence intervals (95% CI). DM risks were compared using Chi-square (significance level p<0.05).
Results: In total 5175 cats were included in the study, of which 65 were diagnosed with DM (PP 1.3%; 95%CI: 0.99-1.61%; 38 males, 27 females, mean weight 4.9±1.2 [SD] kg, age 10.9±3.5 years). The PE-group (n=750) was treated median 2.7 (range 2-4) times/year with progestogens and had DM diagnosed in 21 cases (PP: 2.8%; 95%CI: 1.62-3.98%; OR: 2.89; 95%CI: 1.71-4.90; OR [male]: 1.35; 95%CI: 0.59-3.09; OR [female]: 7.66; 95%CI: 3.57-16.46). The GE-group (n=802) was treated with glucocorticoids during a median 1.6 (range 1-3) episodes prior to consultation and had DM diagnosed in 13 cases (PP: 1.6%; 95%CI: 0.73-2.47%; OR: 1.38; 95%CI: 0.75-2.55%; OR [male]: 1.61; 95%CI: 0.76-3.41; OR [female]: 0.95; 95%CI: 0.32-2.78%).The NPE group (n=4425) had DM diagnosed in 44 cases (PP: 1%; 95%CI: 0.71-1.29%), significantly less than in the PE-group (p<0.001). The NGE (n=4373) contained 52 diabetic cats (PP: 1.1%; 95%CI: 0.79-1.41%), not significantly less than in the GE-group (p=0.31); and the NPGE-group (n=3623) 31 (PP: 0.8%; 95%CI: 0.51-1.09%), significantly less than in GE+PE-groups combined (p<0.001).
Conclusions: Progestogen and/or glucocorticoid administration was documented in more than half of all diabetic patients on record. An 8-fold increased risk for contracting DM after exposure to progestogens was documented in females. Clinicians contemplating the use of such drugs should take account of these results and inform cat owners about this specific diabetogenic risk.
Feline diabetes mellitus (DM) is an important metabolic disorder, with an increasing incidence. In the last years a series of studies were published, each underlying an apparent increase of DM cases, caused by a higher awareness of clinicians, but also a real increase in the DM rate of occurrence. Studies have evaluated the influence of predisposing characteristics in felines and have reported advanced age, male gender, neutered and/or obese individuals and Burmese breed to be favouring factors for diabetes mellitus (McCann et al., 2007; Prahl et al., 2007). Among the favouring/determining factors, insulin antagonistic drugs, such as synthetic progestagens and exogenous glucocorticoids, can be suspected. Although it is not yet well known if this type of drugs hold a real involvement in diabetes mellitus rate of occurrence, both types act as strong insulin antagonists and could be contributing factors in the physiopathogenesis of persistent hyperglycaemia (Behrend and Kemppainen, 1997; McCann et al., 2007; Middleton and Watson, 1985; Peterson, 1987; Scott-Moncrieff, 2010). Progestagens are thought to be involved in DM occurrence trough their reducing effect on the migration of GLUT-4 transporters to the membrane and reducing glucose uptake in the intracellular space. Progestagens also stimulate the secretion of growth hormones, which are still important contributors to insulin resistance (Hoenig et al., 2000). Glucocorticoids were reported to increase hepatic gluconeogenesis, decrease tissue glucose utilization, reduce affinity and number of GLUT-4 glucose transporters and reduce affinity of insulin receptors (Scott-Moncrieff, 2010). As a consequence, in the presence of high concentrations of insulin antagonists and hyperglycaemia, the pancreas usually responds by insulin overproduction and development of islet amyloidosis, to the exhaustion and shutdown of β-cells (Behrend and Kemppainen, 1997; Hoenig et al., 2000; Middleton and Watson, 1985; Zini et al., 2010). Although there are still uncertainties about the promoter of DM, direct drug activity is highly associated with disturbances on insulin secretion and glucose homeostasis (Goossens et al., 1998; Gunn-Moore, 2005; McCann et al., 2007; Scott-Moncrieff, 2010) with predilection in individuals with already impared insulin secretion (Nelson et al., 1999). Thus, both type of drugs could have the capacity to induce disturbances on the insulin production and path-way, (Middleton and Watson, 1985; Peterson, 1987; Scott-Moncrieff, 2010), reduce clearance of intravenous glucose load, decrease glucose excretion rate and cause fasting hyperglycemia (Appleton et al., 2001; Blois et al., 2010; Gottschalk et al., 2011; Middleton and Watson, 1985; Scott-Moncrieff, 2010).
The aim of the current study was to quantify the risk for feline DM associated with use of exogenous progestogens and glucocorticoids in Romania.
Materials and methods
Data was collected from the medical records of cases presented for consultation at the veterinary teaching hospital of Faculty of Veterinary Medicine of Iasi, Romania. Case retrospective registration started with September 2005 and was carried until September 2014. Records of 5175 cats, both male and female, with ages ranging from 4 weeks to 18 years were consulted. Cases were considered eligible for inclusion in the study if they were registered with characteristic clinical signs and a clear diagnosis of DM. The diagnosis was based on the persistent fasting hyperglycemia over 270 mg/dl (80-120 mg/dl), glycosuria, clinical signs of polyuria-polydipsia, polyphagia and weight loss, and data obtained on the general serum biochemistry (Blois et al., 2010; Reusch, 2011). Data recorded for each diabetic cat also included: signalment (age, gender, reproductive integrity, weight and breed), a clear DM diagnosis and exogenous progestogens and/or glucocorticoids (including topicals) exposure. Information regarding progestagen and glucocorticoid administration targeted dosage, number of days per episode and number of episodes per year. Records of non-diabetic cats were used as control and data regarding age, gender, breed, body weight and reproductive integrity were registered as categorical variables.
Univariate analysis were performed for cats included in the study, to identify risk factors associated with diabetes mellitus, for the following variables and groups: gender (male/female), reproductive integrity (neutered/entire), age (<5 years, 5-10 years, 10-15 years, >15 years), body weight (<5 kg, >5kg); breed risk was calculated for cats with 10 or more representative cases of DM and included Burmese, domestic short hair (DSH) and a group of other breeds. Potential risk factors with a P value ≤ 0.1 for univariate analysis and p=≤0.05 for multivariable analyses were considered significant.
Multivariate risk analysis was performed by dividing diabetic cats documented to receive one of the drugs in two groups such as: PE – progestogen-exposed and GE – glucocorticoid-exposed. Data collected from DM cases never exposed to progestagens and/or glucocorticoids were divided in three groups as follows: NPE – no progestogen-exposure; NGE – no glucocorticoid-exposure; NPGE – no progestogen or glucocorticoid-exposure and were used as controls for risk factor analysis.
Period prevalence (PP) and standard logistic regression with odds ratios (OR), as well as 95% confidence intervals (95%CI) were calculated to observe differences in DM occurrence for all groups established for evaluation.
Diabetes mellitus was diagnosed in 65 of the 5175 cats presented during September 2005 to September 2014, revealing a PP of 1.3% and 95%CI of 0.99-1.61%. Of the total DM cases, 38 were males (58.5%) and 27 females (41.5%). A number of 13 males (20%) were intact and 25 (38.4%) neutered. In the female group, 19 (29.3%) were intact and 8 (12.3%) neutered. The average age of the group was 10.9±3.5 years, with ranges from 3 to 18 years. The mean weight of total 65 DM cases was 4.9±1.2 [SD] kg, ranging from 2.8 and 8 kg. Breed distribution comprised 32 Burmese cats (49%), 24 DSH cats (37%) and 9 other breeds (14%) (Table 1).
|Table 1. General characteristics of DM cats|
Progestagen administration has been documented in 21 cats (32.3%), of which 7 were males (10.76%) and 14 females (21.53%) (Table 2). In male cats, progestagens were administered in doses of 2 mg/kg, for 3 to 7 days in male cats. In females, progestagen doses ranged from 2.5 to 5 mg/cat, administered for 3 to 8 days and were aimed to interrupt or delay oestrus. The episodes of administration per year registered an average of 2.7/year, ranging from 2 to 4 series.
Glucocorticoids (methylprednisolone and prednisolone) were administered in both male and female cats for 5 to 21 days, in doses ranging from 0.2 mg to 1 mg/kg as anti-allergic, anti-inflammatory or immunosuppressant. Glucocorticoids were administered on a median of 1.6 (range 1-3) episodes prior to consultation.
|Table 2. Progestagen and glucocorticoid exposure status|
An increase of diabetic PE and GE cats was observed over the 2005-2014 along with the increase of hospital prevalence for the analysed period (Figure 1).
Fig. 1. Hospital prevalence of feline diabetes mellitus increased during the period
September 2005 to September 2014.
(PE Pearson x2 score 0.93 p<0.001; GE Pearson x2 score 0.72 p<0.05)
There was no significant statistical predisposition for male/female groups (χ2= 0.31, p=0.57), nor for DM association with the intact/neutered status in male (χ2= 0.07, p=0.78) or females cats (χ2= 0.41, p=0.51). In age groups, the lowest risk was observed in cats younger than 7 years (PP: 0.5%, 95%CI: 0.17-0.83; OR: 0.27, 95%CI: 0.13-0.58, p<0.0007), followed by cats with ages 7 – 12 (PP%: 1.2%, 95%CI: 0.75-1.65; OR: 0.92, 95%CI: 0.56-1.51, p=0.75); a higher predisposition has been observed for individuals >12 years old, (PP%: 2.5; 95%CI: 1.62-3.38%; OR: 2.88, 95%CI: 1.76 – 4.72, p<0.0001). A significant statistical association has been observed for body weight and DM (χ2= 99.1, p<0.0001), with an increased risk for DM in cats weighing >5kg, (PP%: 3.61, 95%CI: 2.67-4.53; OR: 15.05, 95%CI: 7.43-30.52, p<0.0001), compared to cats <5 (PP%: 0.25, 95%CI: 0.09-0.41%; OR: 0.06, 95%CI: 0.03 – 0.13 p<0.0001). Breed risk analysis revealed a significant association of DM with the Burmese breed, (n=32; 49.2%), represented by 20 males (30.7%) and 12 females (18.4%), indicating a PP: 3.4%, 95%CI: 2.25-4.55%; OR: 4.38, 95%CI: 2.68-7.17 (p<0.0001), considerably higher compared to DSH (PP: 0.85; 95%CI: 0.51-1.19%; OR: 0.49, 95 % CI: 0.29-0.82, p=0.006) and other breeds (PP: 0.63%, 95%CI: 0.22 – 1.04%; OR: 0.42, 95%CI: 0.2-0.85% p=0.016) (Table 3).
|Table 3. Summary of univariate risk analyses|
Total group of PE consisted of 750 cats (269 males; 481 females), of which 21 cats (PP: 2.8%, 95%CI: 1.62-3.98%; OR: 2.86%; 95%CI: 1.69-4.85%, p=0.0001) were diagnosed with diabetes mellitus, of which 7 (10.7%) were males (PP: 2.6, 95%CI: 0.7-4.5%; OR: 2.19%; 95%CI: 0.95-5.03%; p=0.0635) and 14 (21.5%) females (PP: 2.91%, 95%CI: 1.41-4.41%; OR: 4.23%; 95%CI: 1.97-9.06%; p=0.0002).
In the GE group were included 802 cats, with 13 (20%) diabetic cats (PP: 1.62%. 95%CI: 0.75-2.49%; OR: 1.36%; 95%CI: 0.74-2.54%, p=0.31) 9 males (7 intact, 10.76%; 2 neutered, 3.07%; PP: 1.92, 95%CI: 0.68-3.16%; OR 1.58%; 95%CI: 0.74-3.37%; p=0.22) and 4 females (3 intact, 4.6%; 1 neutered, 1.53%; PP: 1.19%, 95%CI: 0.03-2.35%; OR 1.03%; 95%CI: 0.35-3.02%, p=0.94).
In the NPE group were included a number of 4425 cats, of which 44 cases were diagnosed with DM indicating a (PP: 1%, 95%CI: 0.71-1.29%; OR: 1.03%; 95%CI: 0.35-3.02%, p=0.94) significantly less than in the PE-group (p<0.001).
The NGE group contained 4373 cats, with 52 diabetic cats, revealing a PP: 1.18%; 95%CI: 0.86-1.5%.; OR: 0.73%, 95%CI: 0.39-1.34%, p=0.31), not significantly less than in the GE-group (p=0.31).
The NPGE-group, with 3623 cats and 31 DM cases is an important indicator of glucocorticoid and progestagen administration showing a PP: 0.85%, 95%CI: 0.55-1.15%; OR: 0.38%; 95%CI: 0.23-0.62, significantly less than in GE and PE groups together (p=0.001) (Table 4).
|Table 4: Summary of multivariate risk analyses|
To the authors knowledge this is the first retrospective analytic epidemiological report of diabetes mellitus in association with synthetic progestagens and glucocorticoids administration in a well defined population of cats in Romania. The PE and GE groups comprised 52.3% (n=34) of total individuals, compared to NPE, NGE and NPGE represented by 47.7% (n=31) of individuals.
In the present study, individuals treated with progestagens and glucocorticoids registered an average weight of 4.6±1.2 kg, lower than the non-exposed diabetic population, where the average weight was 5.09±1.07 kg. Although in some cats the body weight was close to normal, even small increases of body index were reported to act as risk factors for fasting hyperinsulinemia, impaired glucose tolerance (Rand et al., 2004) and decreased insulin sensitivity by up to 50% (Scott-Moncrieff, 2010). Also average age of exposed groups was 10.5±3.5, compared to non-exposed groups 11.3±3.5. Registered differences could indicate a precocious DM occurrence when exposed to progestagen and glucocorticoids. Still, a sum of underlying conditions that interfere with the insulin pathway, such as chronic medical issues, dental problems, reduced physical activity, obesity, diet (Lederer et al., 2003; Rand et al., 2004) associated with drugs which interfere with insulin pathway and glucose metabolism might spur DM occurrence (Rand, 1999). Also, burmese breed, considered to be an important risk factor (Lederer et al., 2003; Lederer et al., 2009; Rand et al., 1997), was a dominant characteristic of the diabetic population (n=32, 49.2%), with half (n=16) of individuals being framed in the PE or GE groups. This could be an indicator that administration of insulin anthagonists in cases enclosing different risk factors (obesity, neutered males, Burmese breed, >7 years old) could be followed by agravation of insulin resistance, beta cell exhaustion and shutdown (Lederer et al., 2003; Rand et al., 2004). Immediate disturbance on glucose homeostasis was observed in a study developed by Middleton and Watson, demonstrating that administration of prednisolone and megestrol in cats for a period of 8 days, can lead to increased fasting glucose, decreased glucose excretion and reduced glucose tolerance (Middleton and Watson, 1985).
Due to the insulin resistance, hyperglycemiant and deleterious effect exerted on the natural secretion of insulin, (Appleton et al., 2001; McCann et al., 2007; Scott-Moncrieff, 2010) knowing whether or not DM diagnosed cats were treated with such medicine should be one of the main concerns when performing the anamnesis of diabetic cats (McCann et al., 2007; Zini et al., 2010). Diabetic cats with a recent history of progestagen and/or glucocorticoid exposure could enter remmission and clinical signs resolution after treatment cessation (Peterson, 1987).
The higher rate of DM occurence in PE and GE groups, could be a first obvious indicator that progestagens and glucocorticoids might entail an true implication in diabetes mellitus occurrence rate in the studied population. Differences observed in period prevalences, odds ratio and 95% confidence interval of exposed and non-exposed groups could raise awareness in this mather. Despite their negative effect on glucose-insulin homeostasis, both type drugs remain popular among veterinarians. Information regarding their exact associated risk for DM could aid the risk-benefit assessment each clinician needs to make before considering their use.
Progestogen and/or glucocorticoid administration was documented in more than half of all diabetic patients on record. An 8-fold increased risk for DM after exposure to progestogens was documented in females. Clinicians contemplating the use of such drugs could take account of these results and thoroughly evaluate therapy protocol. It would be imperious to educate and inform pet owners about the side effects associated to synthetic progestagens in cats and consider neutering as an alternative to control oestrus characteristic behavior. A better understanding of risk factors could promote early implementation of prevention and control measures for diabetes mellitus and also could bring veterinarians to second thoughts before prescribing a series of drugs. The study could provide framework for future studies regarding glucocorticoid and progestagen associated diabetes mellitus.