Sepsis and the critical cat (Proceedings)

In the 1991 consensus conference, sepsis was defined as evidence of infection and the clinical picture of the systemic inflammatory response syndrome (SIRS; i.e. two of the following clinical criteria: tachycardia, tachypnea, fever or hypothermia, and leukocytosis or leukopenia). Severe sepsis is sepsis with evidence of organ dysfunction and hypotension or hypoperfusion. Septic shock is severe sepsis with refractory hypotension.

Due to lack of specificity, the definition of SIRS leaves much to be desired in both human and veterinary applications. In published SIRS criteria recommended for dogs, the sensitivity ranges from 77-97%, while the specificity is between 64 and 77%, no such study has been performed in cats. It is clear from the table below that almost any patient that comes into the Emergency Room will fulfill at least 2 of criteria.

Table 1

In 2001, there was a second (human) consensus conference, which expanded the criteria for SIRS by recommending the inclusion of physical parameters and biomarkers (the term recently coined to describe measurable factors in blood or biological samples that can be used to identify or stage disease). Many of the parameters and some of these biomarkers may also prove useful in veterinary species. This newer approach, designated as PIRO, incorporates 4 factors in the stratification. The first factor is Predisposition. Although many factors can influence "susceptibility", an area of active research in humans involves predisposing genetic factors. While it has long been recognized that certain breeds or families of people may be more susceptible to diseases individual genetic variations or polymorphisms appear to further contribute to susceptibility. Other predisposing factors may be age, concurrent conditions and gender. The second factor is Infection. Clinically, we recognize that certain bacteria, location of infection, or extent of infection, contribute to the risk of developing sepsis or septic shock. The third factor is the host Response. Our current inability to readily identify and monitor biomarkers is the biggest limitation for this component of the stratification scheme. However if we can determine whether the cat is in an excessively pro-inflammatory state versus immune paralysis, or if there is evidence of adrenal or coagulation dysfunction, we will be better able to chose appropriate and directed interventions. The fourth and final factor is Organ Dysfunction. Even more in cats than in people, the extent of organ dysfunction will negatively influence outcome.

Predisposition

There does not seem to be either a breed or sex predilection for sepsis in cats. Cats with pyothorax were more likely to come from multi-cat households and had a tendency to have a higher incidence of outdoor access. In one study of septic peritonitis, male cats outnumbered female cats. Unlike dogs and humans in which diabetes is reported as a predisposing factor for hepatic abscesses, no such predisposition has been reported in cats.

Infection

In a recent study of activated protein C in people with severe sepsis, the respiratory tract was the source of sepsis in over half of the 1690 patients. The gastrointestinal tract and urinary tract were the other major sources representing 20 and 10% of cases respectively. There are no studies in companion animals that report either the cause of sepsis or outcome in large populations.

In a study of severe sepsis in 29 cats the most common causes of sepsis were

                1. Pyothorax 24%

                2. septic peritonitis 14%

                3. endocarditis 14%

                4. pyelonephritis 7%

                5. osteomyelitis 3%

                6. pyometra 3%

                7. bite wounds 3%

While there is no comparable study in dogs, the most commonly reported cause of sepsis is peritonitis. Reproductive organs (uterus or prostate) are also common sources of sepsis in the dog. In cats compared to dogs it appears that pyothorax may lead to sepsis more frequently than pneumonia. Direct comparison of the response of cats to sepsis versus dogs is challenging. Few studies have provided the same diagnostic criteria or reflect the same basic population; however there are several generalizations that can be made regarding cats with sepsis. The first is that cats are more likely to have a hypodynamic response. In other words, they are likely to be hypothermic, pale, and bradycardic. Anemia is a common finding in septic cats. Hypoalbuminemia is common in septic dogs and cats.

The majority of cases of sepsis are from bacteria of enteric origin, predominantly E. coli. However, in cases of feline pyothorax, Pasteurella and Clostridium predominate and in the limited reports of endocarditis in cats, gram positive bacteria and Bartonella were the reported causes. In some cases of feline sepsis, the source of infection is readily apparent. This is true in cases with severe bite wounds, penetrating trauma, or infected wounds. However, determining the source of infection can be more challenging in other cases. Potential sources of infection in cats include pneumonia, pyothorax, septic peritonitis, septic pancreatitis, pyelonephritis, bacteremia secondary to severe gastrointestinal disease, pyometra, hepatic abscesses, endocarditis, or meningitis.

In dogs and cats with septic peritonitis, the gastrointestinal tract is the most likely source of infection; however, there appear to be some distinct differences. Cats are more likely to have intestinal neoplasia as a cause of peritonitis. Dogs appear more likely than cats to have gastric rupture (ulcers, gastric dilatation and volvus). Dogs are more likely than cats to have reproductive organs as a source of infection. Cats are more likely to have an undiagnosed source of peritonitis than dogs.

Response

Cats are generally more challenging to identify sepsis than dogs. Dogs classically present with signs similar to people, fever, tachypnea, tachycardia and leukocytosis. Cats are frequently afebrile or hypothermic. Severe sepsis in cats can result in bradycardia rather than the expected tachycardia. Abdominal pain was a common physical examination finding in cats with severe sepsis but curiously not consistent in cats with septic peritonitis. Although hepatic abscesses are relatively uncommon, there have been reports describing the clinical features in both dogs and cats. Cats are less likely to be febrile, more likely to be hypothermic, and less likely to have increased liver enzymes.

Biomarkers

Biomarkers of sepsis are an area of active research in human medicine. Several studies have investigated biomarkers in dogs (i.e. protein C, endotoxin, TNF, C-reactive protein); however there are no studies of biomarkers in clinically septic cats. In an experimental study in which cats were given low dose endotoxin, rectal temperature increased, blood pressure decreased, TNF, IL-6, IL-10 and the chemokine CXCL-8 all increased. In addition, white counts decreased, lactate, glucose, creatinine and PT increased. Clinical pathologic findings in cats with sepsis are nonspecific. Leukocytosis is a variable finding in cats with sepsis. Anemia and hypoalbuminemia are typical findings. The pathogenesis of this anemia is complex and multi-factorial. Factors such as frequent blood sampling, nutritional deficiencies, gastrointestinal blood loss, mechanical and antibody mediated hemolysis, and renal or hepatic insufficiency may all play a role. Inflammatory mediators such as tumor necrosis factor – alpha and interleukin-1 inhibit erythroid precursor cells, reduce the formation of erythropoetin (EPO), and blunt the response of the bone marrow to EPO. Feline erythrocytes are particularly susceptible to oxidative damage due to the fact that there are 8 reactive sulfhydryl groups on the hemoglobin molecule.

Decreased plasma ionized calcium concentrations are frequently encountered in human patients with sepsis, and this has also been documented in septic cats. Etiologies identified in humans include parathyroid gland suppression, inadequate vitamin D concentrations, as well as parathyroid hormone – vitamin D resistance, but the cause of these changes in sepsis is still unknown.

The general pattern of glucose derangements in acute sepsis consist of an early and transient hyperglycemia, followed by hypoglycemia. Cats are well known to become hyperglycemic in response to stress, but hypoglycemia is often seen in septic cats. The factors thought to play a role in the development of hypoglycemia include increased peripheral glucose utilization, impaired gluconeogenesis, and depleted glycogen stores.

In cats with severe sepsis, hypoalbuminemia is common. The exact cause is unknown, but hepatic dysfunction, malnutrition and the shifting of protein production from albumin to acute phase proteins all likely contribute. Hepatic dysfunction can be a serious sequela of sepsis. Blood supply to the liver is impaired in septic shock resulting in a decrease in oxygen and nutrient delivery. Hepatic dysfunction can result in coagulopathies, mental depression, hypoalbuminemia and hypoglycemia. Septic cats are often icteric on physical examination. Red blood cell lysis, sepsis induced cholestasis and hepatic dysfunction are all thought to contribute to the development of increased bilirubin in these cats.

Organ dysfunction

The lung is often considered the shock organ in the cat. As such, septic cats may be particularly susceptible to fluid overload, pulmonary edema and pleural effusion are common. This occurs due to increased vascular permeability, sepsis induced myocardial dysfunction, and decreased colloid oncotic pressure due to hypoalbuminemia. This is in contrast to dogs, where the liver and gastrointestinal tract seem to be the primary shock organs.

Approach to the septic cat

Initial diagnostics in these cases generally consist of a complete blood count, serum biochemistry profile, urinalysis and culture, thoracic and abdominal radiographs, and abdominal ultrasound. Additional diagnostics may include blood cultures, endotracheal wash and culture, echocardiography, CSF tap or diagnostic peritoneal lavage

Treatment of sepsis in cats

The goal of treatment in septic cats is the same as in any other patients – maximize perfusion and oxygen delivery to the tissues. Cardiac output is calculated as heart rate x stroke volume. Stroke volume can be improved by maximizing preload with fluid administration. The shock bolus in cats is 50 – 60 ml/kg of crystalloids or 5 ml/kg of colloids. It is recommended that clinicians start with small boluses of 10 – 20 ml/kg given to effect (i.e., normalization of BP and/or CVP).

In cats that remain hypotensive despite adequate volume replacement and normothermia, exogenous catecholamine therapy may be necessary. Dopamine is a vasopressor with positive inotropic, chronotropic and vasoconstrictive effects (5 – 15 mcg/kg/min). This drug is often the first line in septic cats. Dobutamine also exhibits positive inotropic and chronotropic effects and has been shown to improve cardiac output as well as oxygen delivery in experimental models of sepsis. Cats treated with dobutamine should be observed closely, as there are reports of seizures in cats treated with dobutamine CRIs for greater than 24 hours. In cats with severe hypotension, or those that fail to respond to dopamine or dobutamine, norepinephrine (0.1 – 3 mcg/kg/min) or epinephrine (0.1 – 2 mcg/kg/min) are often effective.

Supplemental oxygen is often necessary in these cats. Increased inspired oxygen can improve oxygen saturation (SpO2) as well as the PaO2 thereby improving oxygen delivery. As discussed previously, fluid overload and pulmonary edema are common in these patients. Inflammatory lung disease (acute respiratory distress syndrome) and pneumonia can also contribute to hypoxia and decreased oxygen delivery. In most cats, an oxygen cage or mask oxygen is sufficient, but in severe cases, positive pressure ventilation may be necessary.

Septic cats are often anemic and may be coagulopathic. The administration of red blood cells to an anemic patient will improve the oxygen carrying capacity of the blood and therefore improve oxygen delivery. In addition, replacement of coagulation factors with plasma therapy minimizes further blood loss and provides colloidal support. Red blood cell transfusions should be considered in any septic cat with a PCV < 20, and plasma should be administered in cats with prolonged coagulation times.

Appropriate antibiotic therapy is an integral part of the treatment of septic cats. In humans, inadequate antimicrobial therapy has been associated with increased mortality. Culture and sensitivity should be used to identify effective antibiotics. Broad spectrum antibiotics should be instituted pending sensitivity results.

Nutritional support is extremely important in the septic patient. The systemic inflammatory response in septic cats results in a catabolic state which can rapidly result in severe nutritional deficiency. Nutritional needs should be addressed early in the course of hospitalization. This support can be provided enterally by a naso-esophageal, esophagostomy or PEG tube if the patient is not vomiting and is normothermic. In cats that cannot tolerate enteral feeding, total parenteral nutrition may be necessary. Recent studies in humans have shown that maintaining normoglycemia significantly improves outcome in critically ill patients. We have previously discussed that many of these cats are hypoglycemic on presentation. However, during hospitalization, particularly in patients receiving TPN, the clinician must carefully monitor for hyperglycemia and treat with low doses of insulin when appropriate.

Pain management is also important in these cats. Pain can be difficult to recognize in veterinary patients, particularly in critically ill cats. The severity of illness in these cases often results in reluctance to treat with analgesic agents. Untreated pain in these cats will lead to depression, inappetance/ anorexia, decreased mobility, and an increase in stress hormones. Appropriate use of analgesics is an essential part of appropriate management of the septic cat.

Corticosteroid use in the treatment of sepsis and septic shock has been an area of controversy since the 1950's. Although placebo-controlled trials failed to document beneficial effects of high dose corticosteroids for the treatment of septic shock in people, recent studies have suggested septic human patients may benefit from physiologic doses of steroids. A study evaluating adrenal function in critically ill cats with a variety of disease processes failed to document adrenal insufficiency in any of the patients. (Prittie et al, 2003) It is important to note that the low number of septic cats (n=3) limits the conclusions that may be drawn and warrants further investigation. In humans the current recommendation is the use of replacement doses of glucocorticoids in patients with refractory septic shock. Without further investigation this recommendation cannot be made in our feline patients.

References available upon request