A retrospective evaluation of confirmed and suspected poisonings in 166 cats between 2016 and 2020

Background and Aim: Poisonings commonly bring cats and dogs to veterinary emergency facilities. This retrospective study aimed to analyze clinical signs, confirmed or suspected toxicants, treatments, and outcomes of feline poisoning cases presented over 5 years to the emergency service of a small animal referral center. Materials and Methods: Medical records of 166 cats were evaluated for a history of confirmed or presumed poisoning, suspected toxicant, clinical signs, treatment, and outcome. Poisoning probability was determined using patient history, clinical findings, observation, toxicologic examination, and, in some cases, gastric contents. Results: Most cats were hospitalized (94.0%) due to poisoning with mostly unknown toxicants (48.2%), rodenticides (21.1%), and various toxic plants (12.0%), followed by antiparasitics (6.0%), chemicals (6.0%), drugs (4.2%), tetrahydrocannabinol (1.2%), or inhaled smoke (1.2%). Patients presented predominantly with neurologic deficits (68.7%), reduced general condition (60.2%), and hypothermia (43.4%). The survival rate was 88.6%. Most cats (93.2%) showed no apparent complications at the time of discharge from the hospital. Toxicant-related complications (48.2%) included thermodysregulation (22.9%), central nervous system signs (18.7%), respiratory issues (7.2%), nephrotoxicity (6.0%), gastrointestinal complications (4.8%), evidence of hepatic failure (4.8%), and hemorrhage (1.8%). Conclusion: In this study, the causative toxicant remained unidentified in many cases. Known poisonings were mostly caused by rodenticides. Neurological signs were the most common clinical presentation. Survival rates were high and comparable with those reported by others.

Challenges in recognizing or confirming poisoning in cats are primarily the varying nonspecific clinical signs [13].Therefore, diagnosing poisoning is often based on excluding other differentials.Neurological signs often dominate; however, the clinical sign severity in confirmed or suspected poisoning cases can vary from very mild and barely recognizable to overt presentations with fatal outcomes [18,19].The odds of survival primarily depend on the type and amount of toxicant and the timely initiation of treatment [19][20][21].With few exceptions, any patient with suspected poisoning is to be treated as a medical emergency, requiring an initial clinical assessment followed by stabilization of the patient [21].Supportive therapy, including early decontamination (e.g., bathing, induced emesis, gastric or rectal lavage, and activated charcoal) and elimination, is the cornerstone of treatment [6,7,[21][22][23][24][25][26][27][28][29].In particular, the rising incidence of accidental permethrin application warrants local decontamination measures [6].Accidental drug toxicities are rising among indoor cats; therefore, intravenous (IV) lipid therapy has gained popularity for eliminating these substances [11,[30][31][32][33].If the specific toxicant is known, specific antidote therapy should be used whenever possible and available [25,[34][35][36].
Given the frequency of cats being presented to veterinary emergency centers due to poisonings, providing an overview of the clinical presentation, treatments, and outcomes of confirmed or suspected poisonings in cats is warranted.Therefore, this study aimed to retrospectively evaluate a large cohort of cats with confirmed or suspected poisonings, including outcomes that were presented at the emergency service of a transregional veterinary referral center in Germany over 5 years.

Ethical approval
This study did not involve the use of animals; therefore, ethical approval was not specifically required for this study.

Study period and location
The study was conducted from January 2016 to December 2020 at the Clinic for Small Animals "Tierklinik Hofheim", Hofheim,Germany.

Study design
Electronic medical records (MR) of the emergency service of a large veterinary referral center were searched for "poisoning/intoxication" and "cat" presented between 2016 and 2020.Out of 17,303 cats presented to the emergency service during the same period, 166 cases were identified and included in the retrospective evaluation.
Information retrieved from the MR were: (1) Signalment, (2) level of certainty of toxicant exposure, (3) clinical signs, (4) treatment by the owner or referring veterinarian before presentation, (5) clinical signs and physical examination findings at admission and during hospitalization, (6) the toxicant (confirmed or suspected) and duration from the time of exposure to appearance of clinical signs or admission to the clinic, (7) route of poisoning, and (8) clinicopathologic findings.Routine blood tests included a hematology (ProCyte Dx, Idexx, Kornwestheim, Germany), biochemistry profile (Integra 400 plus, Roche Cobas, Rotkreuz ZG, Switzerland), ionized calcium concentration (I -Stat, Scil, Viernheim, Germany), and clotting times (QuickVet, Scil, Viernheim, Germany).Furthermore, (9) treatment, including decontamination and elimination strategies, (10) outcome, and (11) disease course with any complications and duration of hospitalization were analyzed.For cats treated on an outpatient basis, the duration of treatment was inferred from the MR.
By integrating all available information from individual cats, the suspicion for poisoning was classified as either "certain," "very likely," or "probable." The basis for assigning cats to this classification was, in descending order, "patient history," "clinical signs," "patient observation," "toxicologic examination," and "gastric contents," yielding differentially weighed probabilities of poisoning for the cats included.
Poisonings were classified as "certain" if the ingestion of a specific toxicant was observed; the toxicant was detected on a toxicologic urine analysis (Forensic Toxicology Laboratory and Clinical Toxicology Laboratory, University of Göttingen, Germany) or, in select cases, if clinically and clinicopathologically evident.Cats were classified as "very likely" poisoned if showing typical clinical signs of the suspected poisoning and "probable" if the poisoning was possible and suspected by the attending veterinarian, given the clinical presentation of the cat; however, the patient history was not conclusive.Cats with primary neurological disorders were excluded from the study.In addition, cats were excluded if the owner solely presumed the poisoning without support from patient history, clinical signs, or clinicopathologic findings.

Statistical analysis
Patient data were recorded using a commercial software (Excel, Microsoft Office 2016, Microsoft Corporation, Redmond, WA, USA) and were statistically analyzed (Prism 5, GraphPad, San Diego, CA, USA).Continuous data were evaluated for normal distribution using the D'Agostino and Pearson normality test, based on which summary statistics were reported as either mean ± standard deviation (parametric data) or median and range (nonparametric data).Possible associations between bradycardia and hypothermia at presentation with outcome (survival vs. non-survival) were assessed using Fisher's exact test.Rectal body temperature and heart rates were compared between survivors and non-survivors using the Mann-Whitney U test.p < 0.05 was considered statistically significant.

8).
Most survivors (137/147, 93.2%) had no apparent complications, and more than half of these cats (85/147, 57.8%) were fully recovered at hospital discharge.Marked clinical signs developed or significant improvement failed in ten cats (6.8%), which were discharged early at the owners' request and showed postictal changes (staggering, central blindness or impaired vision, and inability to walk), anorexia, or persistent marked azotemia.
Half of the cats with Lilium species poisoning (n = 3) developed acute kidney injury (AKI).In one of these cats, azotemia was progressive (serum creatinine increased from 1400 to 2365 µmol/L), whereas in the other cats, it was slightly regressive (2095-1833 µmol/L), or only one creatinine   Respiratory rates could not be compared between these groups due to the small number of cats, for which the respiratory rate at presentation was documented (Table -9).

Discussion
This study included 166 cats with confirmed or presumed poisoning that was presented over 5 years.To the best of our knowledge, this reports the largest cohort of feline poisoning cases within a short period [35].
Reasons for hospital admission based on patient history and clinical presentations at hospital admission were largely consistent and comprised mostly neurologic signs or an altered general condition.Neurologic signs are common with poisonings [5,7,12,36].Moreover, witnessing seizures alarms many owners to seek immediate veterinary care.Exhibiting reduced general condition is less specific and can occur with any disease.Cats were also often presented for gastrointestinal signs; however, evidence of altered thermoregulation, cardiovascular, and hydration status predominated on physical examination.A previous review of 138 feline poisoning cases reported similar presentations of neurologic, gastrointestinal and respiratory clinical signs, reduced general condition, and bleeding; however, clinical signs based on medical history were not differentiated from those at admission [35].A possible explanation for more cats showing neurologic clinical signs in this study could be a higher rate of neurotoxic substance intake.However, interpreting clinical signs by medical laypersons versus the veterinary care team and collecting short patient histories during emergency consultations could introduce bias.Furthermore, temporal aspects must be considered as clinical signs may worsen or improve from their onset to hospital admission.
Approximately one-quarter of the cats showed bradycardia or tachycardia, likely due to increased vagal or sympathetic tone associated with the toxicant or specific drug [11,13,35].Tachypnea can also be stress-related or linked to dyspnea.Many hypothermic patients presented intra-or postictally (57.0%) or in advanced shock (28%), and mild hypoglycemia (4.3-4.9 mmol/L) were observed, particularly with rodenticide poisoning (2%).Hypothermia was also detected in cats with comatose states (13%).Only one-third of the hyperthermic cats experienced seizures.More than half of the euthanized cats (56%) and two-thirds of the cats that died were hypothermic.Only one deceased patient was hypothermic.Finally, hypothermia rates were not different between survivors and non-survivors, contrasting with the results of a study that examined one specific toxicant group (organophosphates) [36].Similar to that study, rectal body temperature was lower in non-survivors than in survivors in this study.
Our data indicate that poisonings primarily impact juvenile and young adult cats, possibly attributed to heightened curiosity behavior and associated elevated risk of ingesting foreign substances.However, the causative toxicant could not be determined in almost half of the patients, possibly due to a lower likelihood of witnessing toxicant ingestion in outdoor cats (with a prevalence of 50% in this study) and limited toxicological screenings.Another investigation reported the detection of the toxicant in a larger proportion of patients (80.4%) that comprised only one-third of outdoor cats, and poisonings were significantly more often classified as "certain" than the more conservative classification in our study (72% vs. 23%) [35].Moreover, other studies included cats witnessed to have ingested the toxicant; however, without specifying the level of certainty or subcategories [13,37].Poisonings with ectoparasite prophylactics have been shown to cluster in cats [6,7,18,32,33,[38][39][40].This involves primarily the accidental administration of spot-on formulations that are labeled only for dogs and are highly toxic to cats, where accidental oral administration by the owner or ingestion by the cat during grooming can cause neurotoxicity [38].Two-thirds of the permethrin-poisoned cats in the present study survived (4/6; 67%), and survival rates for other toxicants ranged from 63% to 98% [7,39,40].
Almost a third of all anemic cats (hematocrit [HCT] <30.3%; median age 5.1 years) in our study were affected by poisoning with coumarin derivates, suggesting (mostly occult) bleeding.Furthermore, chronic kidney disease (CKD, n = 2) was considered to be an important contributor to anemia in these cats, and a small deviation (<1%) from the lower reference limit for HCT was interpreted as an age-related physiological finding in juvenile cats (n = 3).Increased serum creatinine concentrations (>167 µmol/L) in about 10% of the cats suggested direct kidney injury (two-thirds of the cats with Lilium species poisoning were azotemic), shock after dehydration, or CKD.Serial creatinine measurements performed in some cats revealed progressive azotemia (8%) or relevant increases in serum creatinine levels within the high-normal range (21%).Survival of Lilium species-poisoned cats with AKI was 50%.The previous study has shown a remarkably good prognosis with up to 100% survival following decontamination and IV infusion within 48 h without AKI development [37].As seen in many cats, hyperglycemia can be explained by the physiological stress response.Hypoglycemia suggests severe cardiovascular compromise (e.g., sepsis) and/or consumption of glycogen stores (e.g., post-seizures); however, preanalytic effects cannot be definitively excluded.Prolongation of PT and aPTT was exclusively detected in coumarin derivative poisoning, whereas an isolated aPTT prolongation was also observed.Possible explanations for the increased aPTT are hepatopathy, inflammatory reaction, or disseminated intravascular coagulation.Significant alterations in laboratory parameters are rarely associated with poisoning.
Decontamination measures were performed less often (35%) than in a recent study [37].As the initial emergency action, decontamination must be carefully balanced against potential risks and contraindications for each patient [21,27,41].Decontamination is unhelpful or contraindicated for remote toxicant ingestion, absence of toxic effects, and in patients with impaired consciousness, seizures, compromised swallowing, or other aspiration pneumonia risk factors.Unlike non-volatile oily substances, emesis should not be induced for foaming, corrosive, sharp objects, or volatile hydrocarbons (e.g., gasoline) to prevent aspiration and severe esophageal injuries [24].Induced emesis is reasonable within 1-2 h of toxicant ingestion; however, little evidence exists to support this narrow window of opportunity, and successful decontamination is described up to 4 h after toxicant ingestion if performed promptly [21,24,27,42].Gastric lavage could be considered if emesis induction has failed or is contraindicated, although the clinical benefit remains controversial [19,21,43,44].Additional IV fluid therapy should be considered in most cats with suspected poisoning and needs to be tailored to the individual toxicant and patient status [41].It can support cardiovascular function and needs to be carefully titrated or even avoided with, for example, toxicants that can cause cerebral edema (e.g., bromethalin) or underlying heart disease.A toxicant-specific antidote was applied only for coumarin derivative poisonings (Vitamin K1).Elimination measures were performed almost twice as often as decontamination attempts, which was mainly IV lipid therapy (Lipofundin 20%, B. Braun, Melsungen, Germany) given to 60% of the cats, particularly to those with α-chloralose, permethrin, other biocides, NSAIDs, or THC poisoning.
Approximately 60% of cats were empirically treated with intravenous lipid emulsion (ILE), as no established protocols existed for managing feline poisoning cases involving hydrophobic agents and severe symptoms at that time.Along these lines, the attending clinician decided to treat, for example, α-chloralose poisoning cases with ILE (log P = 0.99).Particularly in light of possible treatment-associated complications (which also occurred), veterinarians should carefully consider routine ILE administration for hydrophobic toxicants for as long as sufficient evidence is lacking (e.g., log P < 1.0) [44,45].
A survival rate of almost 90% and low mortality rates is consistent with other studies reporting an 86%, 83%, and 85% survival rate, respectively [3,35,36].However, data for comparison are sparse in feline medicine.As most studies distinguish survival versus non-survival, this study differentiated natural death (2%) and euthanasia (10%) as lethal outcomes.Owners' reasons for choosing euthanasia vary, and it is conceivable that some of these cats might have survived with additional therapy.
Toxicant-associated complications were more frequent (48%) than in a recent study with 39 cats (28%) [36].Hypothermia was observed predominantly in rodenticides, hyperthermia due to local inflammation in the oral cavity, and pain triggered by didecyldimethylammonium chloride.Respiratory complications were most frequently noted, likely associated with cardiovascular depression and decreased oxygenation with hemorrhage or anemia.Gastrointestinal signs, including vomiting and superficial tongue lesions, can result from absorption and local effects of the toxicant, which may be prolonged.Nephrotoxicity was primarily associated with the ingestion of Lilium species, evident hemorrhage with coumarin derivatives, and a postictal comatose state with poisonings that did not result in hemorrhage.The hepatic metabolization of several toxicants can contribute to the development of liver failure.
We acknowledge the limitation of this study being retrospective in nature, carrying some risk of bias or lack of information.A standard assessment of the neurological status (e.g., Glasgow coma scale) was not utilized, and respiratory rates were mostly reported if abnormal.Laboratory trends were available from only a few patients.Retrospectively evaluating the likelihood of poisoning is challenging, and toxicological screening was rarely performed.Conclusions from suspected or "unknown" poisonings are generally limited and remain speculative for some clinical signs and treatment options.

Conclusion
This study reviewed 166 feline cases of confirmed or suspected poisoning.Although the causative toxicant remained unidentified in most cases, rodenticides were the most common.Affected cats presented predominantly with neurological signs, reduced general condition, and/or hypothermia.A high survival rate was recorded; however, poisonings can present a life-threatening, serious situation requiring immediate and intensive veterinary care.

Table - 2
: Clinical signs reported in the medical history and at the time of admission in 166 cats with confirmed or suspected poisoning.Refers to the number of cats to which at least one of the clinical signs within that category was seen.Multiple entries for individual cats within a category of clinical signs are possible.CRT=capillary refill time, RR=Respiratory rate *

Table - 3
: Physical examination parameters of 166 cats with confirmed or suspected poisoning at presentation.

Table - 4
: Laboratory findings (hematology, clinical chemistry, and clotting times) in 166 cats with confirmed or suspected poisoning at presentation.Decontamination and elimination measures before and after hospital admission of 166 cats with confirmed or suspected poisoning.Refers to the number of cats to which at least one of the decontamination or elimination methods was undertaken.Multiple entries for individual cats are possible.
measurement was performed (2400 µmol/L).All three cats died, whereas the cats with low-grade azotemia (serum creatinine up to 207 µmol/L) survived.Two-thirds of cats with permethrin poisoning (n = 4) survived.One cat experienced recurrent seizures after weaning from sedation and died.The other cat was euthanized after 34 h of sedation, anesthesia, and ventilation.Bradycardia on presentation was not significantly more common in non-survivors (8/18; 44.4%)

Table - 7
: Classification of the 166 cats presented due to confirmed or suspected poisonings, based on the level of certainty to classify cases as certain, very likely, or probable poisoning.

Table - 8
: Suspected or confirmed toxicants in 166 cats presented with clinical signs or a history of poisoning and the corresponding outcomes.

Table - 9
: Outcome related to physical examination findings at presentation in 166 cats with confirmed or suspected poisoning.
Non-parametric data are presented as median (range).x=No p-value reported due to the limited number of patients, T=Temperature, p=Significance level; statistical significance was set at p < 0.05.