Successful cardiopulmonary resuscitation following cardiac arrest in a hypoglycemic cat

Abstract

This case highlights the successful recovery to discharge of a hypothermic cat in cardiac arrest, with minimal lasting clinical signs. Immediately after resuscitation, the cat was blind and non-ambulatory paraparetic. Within 4 days, the cat became fully ambulatory, but vision loss remains. We believe that the cat’s hypothermia likely contributed to this successful outcome. Other factors which may have played a role in the cat’s recovery were the administration of mannitol and anti-seizure medications.

Key clinical message:

We share learning points regarding re-warming rates for hypothermic patients and the use of Doxapram for stimulation of the central respiratory center.

Less than 50% of dogs and cats which suffer a cardiac arrest in a veterinary hospital have return of spontaneous circulation (ROSC) and approximately 10% survive until discharge from the hospital (1,2). Post-cardiac arrest (PCA) syndrome is believed to be responsible for the low survival rate. This syndrome is the result of ischemic injury caused by sudden disruption in oxygen delivery to the body. Multi-organ failure occurs and if ROSC follows, reperfusion injury can contribute to the ischemic damage already present (1,3).

Mild therapeutic hypothermia (MTH) of 30°C to 32°C is an established treatment following arrest in human patients and it has been shown to improve neurological outcome in dogs following cardiac arrest (4–6).

Case description

A 2.5 kg, 11-year-old, spayed female Siamese cat which had been previously diagnosed with diabetes mellitus, was taken to the Western College of Veterinary Medicine Emergency Clinic after being found unconscious by the owner. The previous day, the dose of glargine insulin was increased from 1 IU to 2 IU, q12h (Lantus 100 Units/mL; Sanofi-Aventis Canada, Laval, Quebec) following a blood glucose measurement of 21.2 mmol/L. The cat had eaten well on the morning of presentation after receiving the increased dose of insulin. Approximately 8 h later, the owner came home to find the cat unconscious and it was brought to the emergency service.

On physical examination, the cat was unresponsive with agonal respiratory movements, and the rectal temperature was too low to be measured by a digital thermometer (< 32.0°C). There was no audible heartbeat on thoracic auscultation and the pupils were dilated with no pupillary light response. The menace and palpebral reflexes were absent.

The trachea was immediately intubated with a cuffed endotracheal tube and attached to a Bain breathing system delivering 100% oxygen. Manual lung ventilation was started (10 breaths/min) and capnography indicated an end-tidal CO₂ of 25 mmHg. Massage of the heart [120 to160 beats/min (bpm)] was begun immediately using the cardiac pump technique. Cardiopulmonary resuscitation (CPR) was performed for 2-minute cycles; followed by a brief pause for assessment of pulse and the electrocardiogram. A 22-G over-the-needle catheter was placed in the cephalic vein and epinephrine (Epinephrine HCl 1 mg/mL; Vétoquinol Canada, Lavaltrie, Quebec), 0.01 mg/kg body weight (BW), IV, was administered, followed by a bolus of atropine (Atropine 0.5 mg/mL; Bimeda, Cambridge, Ontario), 0.04 mg/kg BW, IV. The cat was rewarmed using a warm air circulating device (Bair Hugger 3M, London, Ontario) and heated fluid bags. During resuscitation attempts, a bolus of balanced crystalloid fluids (Normosol-R; Abbott Laboratories, North Chicago, Illinois, USA), 15 mL/kg BW, IV, was given. The blood glucose concentration measured at the time of catheter placement was too low to be detected by a glucometer (AlphaTRAK 2; Zoetis, Kirkland, Quebec). A 3-mL volume of glucose (Dextrose 50%; Valesco Pharmaceuticals, St. Thomas, Ontario), diluted to 12 mL, was given IV over 5 min. Ten minutes after starting CPR, the cat still had no ROSC. On consultation with the owners, the CPR attempt was stopped. However, shortly after stopping CPR, a heartbeat was detected via thoracic auscultation and sinus bradycardia was identified on ECG. The heart rate was 81 bpm and resuscitation was continued; SpO₂ was 94% 25 min after starting CPR.

Poor femoral pulse quality prompted delivery of additional treatments. Another bolus of intravenous fluids (37.5 mL Normosol-R) was given and a dobutamine infusion (Dobutamine HCl 12.5 mg/mL; Sandoz Canada, Boucherville, Quebec), 5 μg/kg BW per minute, IV, was started. Shortly after these interventions, SpO₂ was 99% and pulse quality and blood pressure began improving. The dobutamine infusion rate was slowly reduced until it was not required.

Although there was ROSC, the cat was still apneic. Doxapram (Dopram-V 20 mg/mL; Boehringer Ingelheim, St. Joseph, Missouri, USA), 1 mg/kg BW, IV, was given to stimulate the central respiratory center. The cat still had fixed and dilated pupils. To treat cerebral edema, mannitol (Mannitol injection 25%; Fresenius Kabi Canada, Toronto, Ontario), 0.5 g/kg BW, IV, was given over 20 min. During administration of mannitol, the cat slowly regained consciousness and started breathing. It began moving and the endotracheal tube was removed. As hypercapnia was detected on the capnograph before removal of the endotracheal tube, the cat was anesthetized (Alfaxalone, Alfaxan 10 mg/mL; Abbott Laboratories), 1 mg/kg BW, IV, to allow re-intubation. Following administration of the alfaxalone, generalized, rapid, spastic muscle twitches were observed which were successfully treated with diazepam (Sandoz Canada), 0.5 mg/kg BW, IV. Approximately 15 min later, when normocapnia was observed and the cat’s breathing rate became more appropriate, the trachea was extubated, and the cat was placed into an oxygen chamber with the FIO₂ at 65%.

Two hours after extubation, venous blood sample revealed a pH of 7.1, pCO₂ 69.4 mmHg, and glucose concentration of 28.2 mmol/L. The heart rate was 240 bpm and rectal temperature was still too low to be measured. The warm air circulating device and heated fluid bags were continued until the cat became normothermic, which took 4 h. Overnight, the IV fluid rate was 9 mL/h (Normosol-R with 20 mEq/L potassium chloride). Blood glucose concentration remained elevated when measured every 2 to 4 h. During the night, the cat had 2 more seizure-like episodes, which again were treated with intravenous diazepam. Butorphanol (Torbugesic 10 mg/mL; Zoetis, Kirkland, Quebec), 0.15 mg/kg BW, IV, and another bolus of mannitol (Fresenius Kabi Canada), 0.5 g/kg BW, IV, were given to further control these minor seizure-like episodes. During the next day, the cat had neurological deficits including cortical blindness with an intact PLR and no menace reflex. Mentation was normal, but the cat could not stand or walk and had proprioceptive deficits on all 4 limbs. The facial twitches returned and a diazepam (Sandoz Canada) infusion was started (0.5 mg/kg BW per hour) along with phenobarbital (Phenobarbital Sodium 30 mg/mL; Sandoz Canada), 2 mg/kg BW, IV, q12h. The cat was still deemed to be dependent on oxygen therapy based on pulse oximetry and thus was kept in the oxygen chamber. In the morning, the cat ate wet food when offered, and was fed at regular intervals thereafter. As the cat was oxygen-dependent, oxygen was provided through a mask while thoracic radiographs were taken. An abdominal ultrasound examination showed minor evidence of pancreatitis and hepatomegaly. Follow-up biochemistry revealed a slight elevation in aspartate aminotransferase and urinalysis revealed glucosuria. The venous values of pH and pCO₂ were now normal (7.426 and 39 mmHg, respectively), and oxygenation became adequate on room air. Diazepam was discontinued and no further seizure-like episodes were observed. The cat was still unable to stand but proprioception had improved on all limbs. The cortical blindness remained. Physiotherapy was performed, comprising of daily passive range of motion sessions, assisted standing, rhythmic stabilization, and weight shifting.

On her 4th day of hospitalization, the cat began to walk and phenobarbital (Phenobarb 15 mg; Pendopharm, Division of Pharmascience, Montreal, Quebec), 1 mg/kg BW, PO, q12h, was given. Five days after cardiopulmonary arrest, the cat was discharged from the hospital with phenobarbital for 1 mo. The owners were advised to feed the cat a veterinary approved diabetic diet (Purina DM; Nestlé Purina PetCare Canada, Mississauga, Ontario) until recheck examination in 3 d, but glargine insulin therapy was restarted following results of a blood glucose curve. Thirty months post-arrest, the cat was still alive and some vision had returned as she could follow the path of a laser pointer with her head.

Discussion

Return of spontaneous circulation (ROSC) is not the endpoint of resuscitation from cardiac arrest. For dogs and cats, initial ROSC rates range from 35% to 45%, with survival to discharge rates ranging from 2% to 10% (1,2,7). The combination of PCA syndrome and reperfusion injury is believed to contribute to this overall poor survival. The re-oxygenation of the brain post-arrest likely leads to harmful chemical cascades resulting in interference with Ca²⁺ loading, membrane lipid peroxidation, apoptosis, and cell necrosis (8). Studies in human medicine show adequate neurological recovery post cardiac arrest can be achieved in 11% to 45% of patients (9,10). There is limited information regarding neurological recovery rates in veterinary species. Although the cat had significant neurological recovery, it was left with neurological deficits in the form of near complete loss of vision. We believe several factors played a role in this successful CPR: hypothermia, mannitol, normocapnia, and seizure prophylaxis.

Cortical blindness following general anesthesia in the cat is a documented occurrence (11). The blood supply to the feline brain is almost entirely from the maxillary artery and the feline visual cortex appears to have an increased sensitivity to hypoxia (11). There is an important branch of the maxillary artery (rete mirabile) which lies in close proximity to the temporalis and pterygoid muscles of the skull, which could make it vulnerable to compressive or stretching forces when the mouth is opened (12). Feline maxillary artery blood flow, assessed by a variety of methods, has been compared when the mouth is opened and when the mouth is closed (13). Although the use of mouth gags produced the most alterations in maxillary blood flow, even slight opening of the mouth produced a detectable alteration of flow on magnetic resonance angiography (14). Although the cat in the present report did not have a mouth gag, the mouth was opened while the trachea was intubated. It is uncertain whether the cat’s neurological status could have been improved upon.

Mild therapeutic hypothermia of 30°C to 32°C in humans and in dogs and cats has neuroprotective benefits and is used during cardiopulmonary bypass (5). Experimental studies show less neuronal necrosis in the brains of animals subjected to post-ischemic hypothermia compared with normothermic brains (15). Several studies illustrate a significant improvement in neurologic function and histological evidence of brain damage in hypothermic animals compared to normothermic ones (4–6). The protective action of therapeutic hypothermia is considered to be multifactorial, involving the slowing of destructive enzymatic processes and reduction in cerebral oxygen requirements (16). Additionally, therapeutic hypothermia inhibits lipid peroxidation (17) and minimizes both cerebral edema (18) and intracellular acidosis (19).

The optimal duration of hypothermia and temperature range remain the subject of investigation. Evidence suggests that a delay in the cooling of a patient post arrest could negate the benefits of therapeutic hypothermia (20). As such, it is suggested cooling be started as soon as possible. There are no standardized re-warming rates, although evidence suggests a slow rate (0.25°C/h to 0.5°C/h) is reasonable (21,22). In 1 study, peak brain oxidative stress occurred 16 h after reperfusion post arrest in normothermic animals (23). In this study, regulated hypothermia reduced oxidative stress to baseline values during reperfusion. This shows the need for prolonged periods of hypothermia before re-warming.

The literature describes the intentional cooling of the post-arrest patient, as opposed to the pathological hypothermia of the cat in this report. Furthermore, the temperature of the cat was rapidly (~2°C/h) returned to and maintained, at body temperature which is not in keeping with current recommendations (21). It is uncertain what difference there would have been had MTH been correctly implemented; the hypothermia of the cat on arrival may have played a neuroprotective role in its recovery.

Cerebral edema is a common occurrence post-cardiac arrest. Both hypoxia and the inflammatory response during reperfusion play a role in edema formation. Cerebral edema can result in elevated intracranial pressure (ICP), and lowered cerebral perfusion pressures (CPP). Studies in humans have shown that cerebral edema, high ICP and low CPP have a negative impact on neurological prognosis (24,25). Mannitol is an osmotic diuretic which reduces ICP (26). Although there have been no controlled clinical or experimental trials in any species on the use of mannitol after CPR, clinical experience suggests that mannitol can be efficacious in the management of deteriorating neurologic status (1,2). We believe the administration of mannitol also contributed to the resuscitation of the cat.

Carbon dioxide is a potent vasodilator and hypercapnia causes dilation of cerebral arteries and arterioles and increased blood flow, resulting in increased ICP (27). Hypoventilation of the post-arrest patient can therefore result in increased ICP and negative neurological prognosis (24,25). Conversely, hypocapnia causes constriction of cerebral vessels and decreased blood flow results (27). Hyperventilation of the post-arrest human patient can reduce cerebral blood flow, which may cause or exacerbate cerebral ischemia, thus enhancing rather than reducing secondary brain injury (28). Doxapram also decreases cerebral blood and increases cerebral oxygen consumption and is thus contraindicated in the post-arrest patient (29). There is minimal uniform evidence in dogs and cats with regard to optimal CO₂ levels post arrest and prognosis, but in human medicine normocapnia is associated with lower mortality and American Heart Association Guidelines suggest patient PaCO₂ be maintained high-normal (30,31).

There is no literature documenting the prevalence of post-anoxic seizures in veterinary species. In humans, seizures occur in 5% to 20% of comatose cardiac arrest survivors with or without therapeutic hypothermia (31). The presence of myoclonic seizures after anoxia has been identified as a poor prognostic factor (32). There is a lack of consistent evidence in either human or veterinary medicine with regard to seizure prophylaxis and outcome. An experiment involving the use of prophylactic thiopental (60 mg/kg BW) in the PCA period showed an improvement in mortality rate in cats which received the drug compared to a control group (33). In spite of the effects on mortality, treatment had no effect on the neurologic function of survivors. In a canine experiment involving a bundled therapeutic approach (MTH, thiopental, phenytoin, methylprednisolone), seizure prophylaxis was associated with improved neurologic and overall performance and brain histopathologic damage scores compared to other groups (34). It is uncertain whether the use of seizure therapeutics played a neuroprotective role in this recovery, but it is important to note they were administered.

Due to the number of variables, it is not possible to determine the extent to which any of the treatments contributed to the successful CPR and neurological recovery in this cat. This case highlights the need for more research into the area of hypothermia and neurological recovery following arrest in animals.