Unexpected Hyperkalemia in an Anesthetized Dog

Signalment, History, and Presentation

An 8-year-old female spayed Shetland sheepdog presented to a private veterinary referral center for surgical consultation regarding recurrent left tarsal pad swelling and a chronic draining tract. A review of the patient’s medical record revealed a history of persistent implant-associated infection following a left partial tarsal arthrodesis performed 3 years prior. While the majority of the orthopedic implants had been removed previously, numerous broken screw shafts persisted within the tarsus and metatarsus and were the presumptive source of recurrent infection. A recent aerobic and anaerobic culture revealed a Serratia marcescens infection, which was sensitive to and treated with enrofloxacin. The most recent blood work (CBC and serum biochemical profile) was within normal limits, except for an elevated alkaline phosphatase of 276 U/L (reference range, 23 to 212 U/L). The patient was currently being administered carprofen (2.6 mg/kg PO q12h) as an analgesic and anti-inflammatory and a cranberry supplement once daily for long-term urinary tract health. No other clinical signs, diseases, or comorbidities were noted in the record.

Initial Assessment

On physical examination, the patient was quiet, alert, and responsive with the following vital signs:

• Heart rate (HR): 114 beats per minute (bpm)
• Respiratory rate (RR): 24 breaths/min
• Temperature: 37.9 °C (100.3 °F)
• Mucous membranes: Pink
• Capillary refill time (CRT): < 2 seconds
• Body weight: 14 kg (31 lb)
• Body condition score (BCS): 7/9
• Pain score: 1/4

A single lateral radiograph of the left tarsus revealed metallic implant fragments (broken screw shafts) in the proximal and distal tarsus/metatarsus.

After consultation, the client elected to pursue surgical amputation of the left hindlimb to permanently remove the source of infection. Surgery was scheduled for the following day, and presurgery pharmaceuticals were dispensed to the owner (maropitant, gabapentin, and trazodone were to be administered at home on the morning of surgery).

Patient Management

Anesthesia and Surgery

Given the patient’s BCS of 7/9, all anesthetic and analgesic drugs were based on a lean body weight of 12 kg (26 lb; 15% reduction). On the morning of surgery, the owners administered maropitant (2 mg/kg), gabapentin (10 mg/kg), trazodone (5 mg/kg), and carprofen (2.6 mg/kg) orally along with a small breakfast. Upon arrival at the veterinary clinic, the patient’s mentation was noted to be quiet, alert, and responsive. Preoperative vital signs were as follows:

• HR: 72 bpm
• RR: 24 breaths/min
• Temperature: 38.4 °C (101.2 °F)
• Mucous membranes: Pink
• CRT: < 2 seconds

The patient was premedicated with butorphanol (0.3 mg/kg IM) and dexmedetomidine (5 µg/kg IM). Within 15 minutes, the patient was laterally recumbent in her kennel. The patient was placed on a padded gurney and transported to the surgery department. A 22-gauge catheter was placed in her right cephalic vein. During catheter placement, peripheral oxygen saturation (Spo₂), noninvasive oscillometric blood pressure, and electrocardiogram (ECG) were monitored. During that time, the patient received flow-by oxygen supplementation (3 L/min via face mask). The patient’s eyes were lubricated to prevent corneal drying and ulceration. Preoperative blood work (CBC and serum biochemical profile) was performed, which revealed the following abnormalities:

• Alanine transaminase: 183 U/L (reference range, 10 to 125 U/L)
• Total bilirubin: 2 mg/dL (reference range, 0 to 0.9 mg/dL)
• Potassium: 6.9 mmol/L (reference range, 3.5 to 5.8 mmol/L)

A second blood draw and analysis were performed, which confirmed the presence of hyperkalemia (potassium 6.3 mmol/L). During this time, the patient’s vital signs were within normal limits. Due to potassium’s role in the cardiac conduction cycle, particular attention was placed on HR (60 to 80 bpm), blood pressure (mean arterial pressure [MAP] 80 to 92 mm Hg), and ECG (normal sinus rhythm). The medical record was reviewed, and historic potassium levels were confirmed to be within normal limits. After consultation with multiple members of the veterinary team, the attending surgeon decided to proceed to general anesthesia and surgery. Common treatment options for patients with hyperkalemia include IV fluid therapy, IV dextrose, terbutaline, regular insulin with IV dextrose, calcium gluconate, and/or sodium bicarbonate.¹ Therefore, all of the aforementioned medications were verified to be on hand. In addition, serial electrolyte analysis would be performed every 30 minutes throughout the procedure.

General anesthesia was induced by administering ketamine (1 mg/kg IV) followed by a titration of propofol (1.2 mg/kg IV). Endotracheal intubation was performed with a 6.5-mm endotracheal tube; correct tube placement was verified using direct visualization and capnography (end-tidal carbon dioxide [ETco₂] 48 mm Hg). General anesthesia was maintained with isoflurane (initial vaporizer setting of 1.25%) in 100% oxygen via an adult rebreathing circuit with a 1-L reservoir bag. An initial oxygen flow rate of 1 L/min (83 mL/kg/min) was used to rapidly achieve appropriate anesthesia depth. Fluid therapy was initiated using lactated Ringer’s solution at 5 mL/kg/hr IV. A lumbosacral epidural injection was sterilely performed using a combination of ropivacaine 0.5 mg/kg and buprenorphine 4 µg/kg diluted with saline to a final volume of 0.2 mL/kg. The patient was rotated to lateral recumbency in order to perform surgical preparation of the affected limb. The patient was transported to the operating room. Hydromorphone (0.05 mg/kg) IV was administered for primary systemic analgesia. A ketamine constant-rate infusion (CRI) was started (0.5 mg/kg/hr). Cefazolin (22 mg/kg IV) was administered slowly over 10 minutes.

The initial surgical incision was made, and anesthesia depth was deemed appropriate based on cranial reflex assessment (absence of palpebral blink, loose jaw tone, and ventromedial eye position). Due to the absence of observed nociception and use of a multimodal analgesia plan, the isoflurane vaporizer setting was reduced to 1%, and 5 minutes later, the oxygen flow rate was reduced to 0.5 L/min (42 mL/kg/min). Core vital signs throughout the perioperative period were largely within normal limits:

• HR: 50 to 80 bpm
• MAP: 70 to 90 mm Hg
• ECG: normal sinus rhythm
• SPo₂: 95% to 98%
• Temperature: 37.3 °C to 37.8 °C (99.1 °F to 100.1 °F).

Forty-five minutes after induction, mild elevations in ETco₂ were noted (55 to 60 mm Hg) with a RR of 8 breaths/min. This abnormality prompted the use of mechanical ventilation with the following settings: RR 20 breaths/min, tidal volume 150 mL, and peak inspiratory pressure 18 cm H₂O. Following initiation of mechanical ventilation, ETco₂ normalized (40 to 45 mm Hg) and remained so for the remainder of the procedure.

Serial electrolyte analysis throughout the intraoperative procedure revealed the following potassium values: 7.3, 7.5, and 6.6 mmol/L. Although these values are significantly elevated, the patient remained asymptomatic. More specifically, blood pressure remained within normal limits, no cardiac arrhythmias were detected, and no ECG morphology changes were appreciated (e.g., T wave abnormalities, absent P waves, atrial standstill, ventricular beats). While the patient’s heart rate was on the lower side, the degree of bradycardia was not unexpected given the anesthetic drug protocol utilized.

Ultimately, a left hindlimb amputation was successfully performed. Total surgical time was 61 minutes. Total anesthetic time was 124 minutes.

Recovery and Outcome

After surgery, the patient was weaned off mechanical ventilation (spontaneous respiration with a RR of 18 breaths/min and ETco₂ of 42 mm Hg) and then transported to a recovery kennel. Twenty-five minutes after the discontinuation of inhalant anesthesia, vital signs were within normal limits (temperature 37.4 °C [99.4 °F], HR 87 bpm, RR 24 breaths/min, and SPo₂ 96%), but the patient remained moderately sedate. Therefore, atipamezole (0.04 mg/kg IM) was administered. Within 10 minutes, the patient’s level of consciousness improved.

A clean ice pack was applied to the incision for 15 minutes. Lactated Ringer’s solution was reduced to 2.5 mL/kg/hr (a total of 172 mL was delivered during the intraoperative period). Postoperative vital signs were taken every hour for the next 3 hours; no abnormal parameters were observed. Recheck blood work 4 hours later was within normal limits, including a normal potassium level of 4.6 mmol/L. Oral administration of gabapentin (10 mg/kg), trazodone (5 mg/kg), and carprofen (2.6 mg/kg) resumed 4 hours after extubation. Clavamox (14 mg/kg PO) was started that evening. Starting 4 hours after extubation, small amounts of wet food (the exact amounts not quantified) were offered by bowl every 4 hours. Food intake overnight was initially described as “decreased” but progressed to “normal” at the second and third feeding times. The treatment sheet also included orders for hydromorphone (0.05 mg/kg IV) in the event of elevated pain scores and acepromazine (0.01 mg/kg IV) in the event of anxiety; however, neither of these medications were deemed to be necessary overnight (pain scores between 0 and 1 out of 4).

Ultimately, the patient was discharged from the hospital 20 hours after surgery without complication. Postoperative at-home care instructions included gabapentin (10 mg/kg PO q8h), trazodone (5 mg/kg PO q8h), carprofen (2.6 mg/kg PO q12h), cold packs on incision (q6h), and clavamox (14 mg/kg PO q12h).

Ten days after initial surgery, the patient presented for a sedated evaluation of the surgical site. Current medications included clavamox, enrofloxacin, carprofen, gabapentin, trazodone, cranberry supplement, and maropitant. As a precaution, electrolytes were checked prior to sedation (potassium 4.7 mmol/L). The patient was then sedated with acepromazine (0.02 mg/kg IM), alfaxalone (1 mg/kg IM), ketamine (1 mg/kg IM), and methadone (0.3 mg/kg IM). Under sedation, the surgical site was evaluated and a small area of superficial incisional dehiscence was repaired. Intraoperative and postoperative potassium levels were within normal limits (5.4 and 4.1 mmol/L, respectively). The patient recovered uneventfully and was discharged 4 hours later.

Discussion

Hyperkalemia is defined as an elevation in serum potassium levels. In dogs and cats, a potassium level greater than 5.5 mmol/L is considered abnormal. Generally speaking, moderate elevations (> 6 mmol/L) and severe elevations (> 7 mmol/L) may be cause for concern. The potential impact of hyperkalemia on an anesthetized patient is primarily cardiac in nature. More specifically, elevated potassium levels can disrupt cardiac function. The anesthetist may see 1 or more of the following signs in a symptomatic patient: tall T waves, tented T waves, deeper/more negative T waves, absent P waves, atrial standstill, wide and bizarre QRS complexes, bradycardia, hypotension, and/or asystole (FIGURES 1 AND 2). Given its impact on cardiovascular function, recognition and prompt treatment of hyperkalemia in the symptomatic patient are of fundamental importance.

FIGURE 1. ECG tracing demonstrating atrial standstill in a 9-year-old mixed-breed dog with Addison’s disease and hyperkalemia (9.7 mmol/L). Note the absence of P waves preceding each QRS complex. Courtesy Julia Treseder, MS, DVM, DACVIM (Cardiology).

FIGURE 2. ECG tracing demonstrating abnormal T-wave morphology in a 3-year-old cat with urinary obstruction and hyperkalemia (7.8 mmol/L). Note the tall and tented appearance of the T waves. Courtesy Kristin Smith, DVM, DACVECC.

Causes of hyperkalemia can broadly be placed into 1 of 3 categories: (1) increased potassium input, (2) decreased potassium excretion, or (3) translocation from intracellular to extracellular spaces. A review of the available veterinary-specific literature produced several case reports of hyperkalemia in dogs during the perioperative period. In 1 case study, intraoperative hyperkalemia was most likely attributable to propofol administration in combination with poorly controlled diabetes mellitus.² At least 2 reports suggest that certain breeds (namely greyhounds) may be predisposed to developing hyperkalemia.^3,4 Finally, a retrospective analysis published in 2023 found a link between hyperkalemia and the preanesthetic administration of dexmedetomidine plus an opioid.⁵

If one broadens the search criteria to include reports of hyperkalemia in anesthetized patients of all species, the list of contributing factors that may predispose a patient to developing intraoperative hyperkalemia increases dramatically (BOX 1).^2-8

Although the topic of hyperkalemia in the perioperative period is relatively well documented in the literature, this case study remains important for several reasons. First, despite the aforementioned reports, causation of intraoperative hyperkalemia remains elusive. Therefore, additional instances of hyperkalemia must be reported and examined to clarify the nature of this issue. Furthermore, this case study is unique in that many of the previously reported potential contributing factors appear to be unlikely. For example, the first documented instance of hyperkalemia in this patient occurred approximately 20 minutes after premedication with butorphanol and dexmedetomidine. As such, hyperkalemia occurred much more rapidly in this case than in prior reports.^2-5 In addition, the number of medications administered to this patient (i.e., 2 injectable premedicants) was relatively small compared to previous reports.^2-5 Although the patient had received multiple oral medications (gabapentin, trazodone, maropitant, cranberry supplement, and carprofen) within the 24 hours leading up to general anesthesia, it is unlikely that these oral medications caused/contributed to the development of hyperkalemia for the following reasons:

• None of these medications (including the cranberry supplement) contain significant amounts of potassium
• Prior blood work analysis performed while the patient was taking these oral medications revealed normal potassium levels
• Blood work analysis performed 10 days postoperatively while the patient was taking these same oral medications revealed normal potassium levels.

These facts, along with the patient signalment and the lack of preoperative comorbidities, serve to considerably narrow the differential list.

Simply stated, the patient described in this case study most likely experienced hyperkalemia as a result of translocation following the administration of 1 or both of the injectable premedicants (i.e., dexmedetomidine, butorphanol). Of those 2 medications, dexmedetomidine is the more likely culprit. As previously discussed, prior case studies have found a link between the administration of α₂ agonists and elevated serum potassium levels.⁵ One proposed explanation of this effect focuses on α₂ agonists’ inhibitory effect on circulating insulin levels. In turn, decreased insulin leads to elevated blood glucose levels and may impact potassium homeostasis.⁷

The implications of this case study are both valuable and far reaching. The development of hyperkalemia in the perioperative period is well documented. The list of potential contributing factors is both long and diverse. Hyperkalemia may have negative and profound effects on a patient’s cardiovascular system. Therefore, recognition and treatment of hyperkalemia in the symptomatic patient is of paramount importance.