Introduction

There is considerable evidence to support the adjustment of drug dosages in human patients with heart failure, hepatic failure or renal insufficiency. In contrast, similar studies are lacking in dogs and cats. This presentation will discuss veterinary situations in which drug dose adjustments may be warranted.

Heart Failure

Dogs and cats with heart failure have decreased cardiac output, which leads to preferential shunting of blood to the brain and heart, and decreased perfusion of the skin, gastrointestinal tract and kidneys. Therefore, oral drugs may have decreased bioavailability in animals with fulminant failure, particularly when ascites is present. Parenteral dosing should be used whenever possible in these patients.

In patients taking digoxin, increased perfusion of the heart and brain, as well as decreased renal blood flow, will increase the risk of digoxin toxicity, including arrhythmias and centrally mediated nausea. Prerenal azotaemia from low cardiac output will also decrease the clearance of other cardiac drugs, to include enalapril (approximate doubling of plasma concentrations with moderate azotaemia) and furosemide (60% renal clearance in dogs, and also potentiates prerenal azotaemia). Benazepril clearance is not affected by moderate azotaemia in dogs and cats, and is a good option for patients with heart disease and prerenal azotaemia.

Cardiac patients are at risk for many potential drug interactions (Figure 1), because they are typically treated with more than one drug, and many cardiac drugs have a narrow safety margin.

Figure 1. Potential cardiac drug interactions.

Hepatic Insufficiency

Liver disease can be associated with impaired metabolism of some drugs, but this does not typically occur unless there is substantial loss of hepatic dysfunction, i.e., cirrhosis. In humans with inflammatory liver disease without cirrhosis, hepatic drug metabolism is fairly well conserved. Veterinary diseases with substantial hepatic dysfunction include moderate to severe hepatic lipidosis, acute hepatic necrosis and progression to cirrhosis (seen in dogs). In addition, portosystemic shunts will decrease first-pass delivery of drugs to the liver, leading to higher plasma drug concentrations at standard dosages.

Based on human data, dosages of the following drugs should probably be reduced in dogs and cats with severe hepatic dysfunction: propranolol (decrease dose by 50% or more); diazepam, midazolam, and phenobarbital (use 25–50% of regular dose, and use sparingly if treating encephalopathic seizures); and metronidazole (use 25–50% of anti-anaerobic dose).

Portal hypertension and ascites can develop in dogs with progressive liver fibrosis. Lipid soluble drugs such as propofol, fentanyl and vitamin K₁, will not distribute to ascites, and should be dosed based on the normal body weight, minus estimated ascites fluid weight. For relatively water-soluble drugs, such as ampicillin, aminoglycosides and fluoroquinolones, the total bodyweight (including ascites fluid) can be used.

Patients with hepatic insufficiency are reported to have increased sensitivity to central nervous system (CNS) depressants. This may be due to the fact that sedatives, opioid analgesics and injectable anaesthetics are typically lipophilic and therefore require hepatic metabolism in order to be eliminated, but could also be due to direct effects of hepatic failure on brain chemistry. Titrated dosages of reversible opioids, such as fentanyl, oxymorphone or hydromorphone, are good choices in these patients. Non-reversible drugs such as benzodiazepines, acepromazine and phenobarbital should be started at about 25% of standard dosages and titrated upwards if needed.

Hepatic encephalopathy (Figure 2) is an important complication of portosystemic shunts, hepatic lipidosis, acute liver intoxications and progression to cirrhosis.

Figure 2. Factors that worsen hepatic encephalopathy in dogs and cats.

Renal Failure

There are very few studies supporting drug dosage adjustments for dogs or cats in renal failure. Creatinine clearance is used to make rational dosage adjustments in azotaemic humans, but these measurements are typically not available for most veterinary patients. Human drug dosage adjustments are typically made when the creatinine clearances reaches 0.7–1.2 ml/min/kg, depending on the drug. This corresponds to a serum creatinine of about 221–309 µmol/l in dogs and cats (2.5–3.5 mg/dl). Based on data generated in humans with renal disease, veterinary nephrologists at the annual Renal Therapies Symposium at the Animal Medical Center have recommended dosage adjustments for dogs and cats based on IRIS (International Renal Interest Society) stage. These expert opinion guidelines are listed in Figure 3, and are the best that we have until more studies are done.

While beta lactams are renally cleared, they have high margins of safety and are inexpensive, and, under most clinical situations, probably do not require dose adjustments. However, most fluoroquinolones are renally cleared and can cause dose-dependent cartilage damage in growing dogs and dose-dependent retinal toxicity in cats. Given the high prevalence of renal insufficiency in older cats, less retinotoxic fluoroquinolones should be selected, such as marbofloxacin and orbifloxacin, and label dosages should never be exceeded. Adjustment of both dose and dosing interval has been recommended for fluoroquinolones in humans.

Aminoglycosides are dose-dependent nephrotoxins that cause cumulative renal damage with duration of treatment. For resistant gram-negative infections in patients with renal insufficiency, other drugs should be chosen whenever possible. Cefotetan can be given subcutaneously at home (30 mg/kg s.c. q12h). Meropenem can also be given at home (8–12 mg/kg s.c. q12h) but vials are only stable in the refrigerator for 24 hours, and the drug is quite expensive. Ticarcillin and aztreonem require hospitalisation for intravenous administration.

If aminoglycosides must be used, always rehydrate first and use concurrent fluid therapy (intravenous or subcutaneous) during treatment. Aminoglycoside dosing should be adjusted by giving the drug less frequently (extended dosing interval). Whenever possible, trough serum drug concentrations (just prior to the next dose) should be measured and maintained at < 2 µg/ml. It is important to monitor for tubular damage by examining daily fresh urine sediments for granular casts. Aminoglycosides should not be used in patients with urinary obstruction, or concurrently with furosemide or NSAID treatment.

Furosemide alone should be used with caution in renal failure, since it can lead to dehydration, hypokalaemia and occasional acute renal decompensation. In patients with both heart failure and renal insufficiency, furosemide dosages should be titrated to the lowest dose that controls pulmonary oedema, and patients should be frequently monitored for dehydration and electrolyte changes.

All angiotensin-converting enzyme (ACE) inhibitors have potential adverse effects on glomerular filtration rate (GFR), via efferent arteriole vasodilation. This can lead to renal decompensation at high ACE inhibitor dosages combined with either furosemide or NSAID co-administration. Benazepril may be preferred over enalapril in azotaemic patients, since benazepril undergoes some hepatic clearance, and does not require dosage adjustment in mild to moderate azotaemia in cats and dogs (creatinine values up to around 247.5 µmol/l (2.8 mg/dl)). As with furosemide treatment, patients on ACE inhibitors need to be monitored at least quarterly for changes in blood urea nitrogen (BUN), creatinine and serum potassium.

Finally, all NSAIDs have the potential to cause renal decompensation in an at-risk patient. NSAIDs block protective vasodilatory renal prostaglandins that are elaborated in response to decreased renal blood flow and renin-angiotensin activation. This blocks the patient's ability to compensate for nephron loss or dehydration. In azotaemic patients with osteoarthritis, consider alternatives to NSAIDs, such as tramadol, buprenorphine, physical therapy and acupuncture.

If NSAIDs are required for pain control and quality of life, use conservative NSAID dosages and consider supplemental subcutaneous fluids at home. Patients should be monitored at home for inappetence, vomiting or diarrhoea, and clinic rechecks should include an evaluation of BUN and creatinine. Note that coxibs have the same potential to adversely affect GFR as do older NSAIDs, since COX-2 is constituitively expressed in the kidney. Coxibs are therefore not safer than classical NSAIDs in renal insufficiency.

Figure 3. Drug dose adjustment in renal failure (adapted from the Advanced Renal Therapies Symposium at the Animal Medical Center).