Deborah C. Silverstein

Association between outcome and organ system dysfunction in dogs with sepsis: 114 cases (2003-2007)

OBJECTIVE To determine whether multiple organ dysfunction syndrome (MODS) could be identified in dogs with sepsis secondary to gastrointestinal tract leakage, and whether the number of affected organ systems was significantly associated with mortality rate. DESIGN Multicenter retrospective case series. ANIMALS 114 dogs. PROCEDURES Medical records for dogs treated surgically because of sepsis secondary to gastrointestinal tract leakage between 2003 and 2007 were reviewed. Sepsis was diagnosed on the basis of results of bacterial culture of peritoneal fluid, gross evidence of gastrointestinal tract leakage at surgery, or both. Renal dysfunction was defined as a > or = 0.5 mg/dL increase in serum creatinine concentration after surgery. Cardiovascular dysfunction was defined as hypotension requiring vasopressor treatment. Respiratory dysfunction was defined as a need for supplemental oxygen administration or mechanical ventilation. Hepatic dysfunction was defined as a serum bilirubin concentration > 0.5 mg/dL. Dysfunction of coagulation was defined as prolonged prothrombin time, prolonged partial thromboplastin time, or platelet count < or = 100,000/microL. RESULTS 89 (78%) dogs had dysfunction of 1 or more organ systems, and 57 (50%) dogs had MODS. Mortality rate increased as the number of dysfunctional organ systems increased. Mortality rate was 70% (40/57) for dogs with MODS and 25% (14/57) for dogs without. CONCLUSIONS AND CLINICAL RELEVANCE Results indicated that MODS, defined as dysfunction of at least 2 organ systems, can be identified in dogs with sepsis and that organ system dysfunction increased the odds of death.

Early nutritional support is associated with decreased length of hospitalization in dogs with septic peritonitis: A retrospective study of 45 cases (2000-2009).

OBJECTIVE To determine whether the timing and route of nutritional support strategy affect length of hospitalization in dogs with naturally occurring septic peritonitis. DESIGN Retrospective study encompassing cases from 2000 to 2009. SETTING University teaching hospital. ANIMALS Forty-five dogs that survived septic peritonitis. INTERVENTIONS None. MEASUREMENTS AND MAIN RESULTS Nutritional strategy for each dog was categorized as either enteral nutrition (EN: free choice voluntary eating or assisted tube feeding) or central parenteral nutrition (CPN). Early nutritional support was defined as consistent caloric intake initiated within 24 hours postoperatively. Consistent caloric intake occurring after 24 hours was defined as delayed nutritional support. Data reflective of nutritional status included body condition score, serum albumin concentration, and duration of inappetence before and during hospitalization. Body weight change from the beginning to the end of hospitalization was calculated. A modified Survival Prediction Index 2 score was calculated for each dog at admission. Additional clinical data recorded for comparison of illness severity included indicators of severe inflammation (eg, presence of toxic changes in neutrophils and immature neutrophils), coagulopathy (eg, prolonged prothrombin time and activated partial thromboplastin time), the use of vasopressors and blood transfusions, and presence of concurrent illnesses. Nutrition-related complications were classified as mechanical, metabolic, or septic complications. Multivariate linear regression analyses were used to determine the relationship of nutritional strategy with hospitalization length, while considering the presence of nutrition-related complications, the nutritional status- and illness severity-related variables. While controlling for other variables, dogs that received early nutrition had significantly shorter hospitalization length (by 1.6 days). No statistically significant association was found between route of nutrition and hospitalization length. The presence of concurrent illnesses and nutrition-related metabolic complications were also associated with longer hospitalization length (by 2.1 and 2.4 days, respectively). CONCLUSION Early nutritional support in dogs with septic peritonitis is associated with a shorter hospitalization length.Objective To determine whether the timing and route of nutritional support strategy affect length of hospitalization in dogs with naturally occurring septic peritonitis. Design Retrospective study encompassing cases from 2000 to 2009. Setting University teaching hospital. Animals Forty-five dogs that survived septic peritonitis. Interventions None. Measurements and Main Results Nutritional strategy for each dog was categorized as either enteral nutrition (EN: free choice voluntary eating or assisted tube feeding) or central parenteral nutrition (CPN). Early nutritional support was defined as consistent caloric intake initiated within 24 hours postoperatively. Consistent caloric intake occurring after 24 hours was defined as delayed nutritional support. Data reflective of nutritional status included body condition score, serum albumin concentration, and duration of inappetence before and during hospitalization. Body weight change from the beginning to the end of hospitalization was calculated. A modified Survival Prediction Index 2 score was calculated for each dog at admission. Additional clinical data recorded for comparison of illness severity included indicators of severe inflammation (eg, presence of toxic changes in neutrophils and immature neutrophils), coagulopathy (eg, prolonged prothrombin time and activated partial thromboplastin time), the use of vasopressors and blood transfusions, and presence of concurrent illnesses. Nutrition-related complications were classified as mechanical, metabolic, or septic complications. Multivariate linear regression analyses were used to determine the relationship of nutritional strategy with hospitalization length, while considering the presence of nutrition-related complications, the nutritional status- and illness severity-related variables. While controlling for other variables, dogs that received early nutrition had significantly shorter hospitalization length (by 1.6 days). No statistically significant association was found between route of nutrition and hospitalization length. The presence of concurrent illnesses and nutrition-related metabolic complications were also associated with longer hospitalization length (by 2.1 and 2.4 days, respectively). Conclusion Early nutritional support in dogs with septic peritonitis is associated with a shorter hospitalization length.

Effectiveness of intravenous fluid resuscitation in the emergency room for treatment of hypotension in dogs: 35 cases (2000–2010)

OBJECTIVE To assess the effectiveness of fluid resuscitation for the treatment of hypotensive dogs presented to the emergency room. DESIGN Retrospective study (2000-2010). SETTING University teaching hospital. ANIMALS Thirty-five dogs with confirmed hypotension. INTERVENTIONS None. MEASUREMENTS AND MAIN RESULTS Medical records from 2000 to 2010 were searched for dogs that had documented arterial hypotension (Doppler blood pressure <90 mm Hg) upon presentation to the emergency room. The following data were collected for retrospective analysis: signalment, body weight, systolic blood pressure (BP) and heart rate (HR) before and after fluid resuscitation, type and volume of fluid administered within the first hour of hospitalization, presence or absence of confirmed sepsis, and outcome. There was a significant increase in BP following fluid resuscitation and 23 dogs showed normalization of blood pressure (Doppler blood pressure >90 mm Hg) following bolus IV fluid therapy within 1 hour of presentation. Fluid responsive animals were significantly less likely to be euthanized compared to those animals that did not respond to a fluid bolus (P = 0.013). The HR did not change significantly in either group postbolus therapy. Twenty-three dogs (65.7%) in the study received only isotonic crystalloid therapy, 11 dogs were administered both isotonic crystalloids and synthetic colloids, and 1 dog received only synthetic colloid resuscitation. A significant difference was not detected between groups with respect to type or volume of fluid administered. CONCLUSIONS Bolus fluid therapy for the treatment of hypotensive dogs resulted in increased in systolic arterial BP in all dogs, although the HR did not reliably decrease as might be expected. Dogs that showed normalization of BP within the first hour of fluid resuscitation were more likely to be discharged alive than those who remained hypotensive. A majority of dogs received and responded to isotonic crystalloids fluid resuscitation.Objective To assess the effectiveness of fluid resuscitation for the treatment of hypotensive dogs presented to the emergency room. Design Retrospective study (2000–2010). Setting University teaching hospital. Animals Thirty-five dogs with confirmed hypotension. Interventions None. Measurements and Main Results Medical records from 2000 to 2010 were searched for dogs that had documented arterial hypotension (Doppler blood pressure 90 mm Hg) following bolus IV fluid therapy within 1 hour of presentation. Fluid responsive animals were significantly less likely to be euthanized compared to those animals that did not respond to a fluid bolus (P = 0.013). The HR did not change significantly in either group postbolus therapy. Twenty-three dogs (65.7%) in the study received only isotonic crystalloid therapy, 11 dogs were administered both isotonic crystalloids and synthetic colloids, and 1 dog received only synthetic colloid resuscitation. A significant difference was not detected between groups with respect to type or volume of fluid administered. Conclusions Bolus fluid therapy for the treatment of hypotensive dogs resulted in increased in systolic arterial BP in all dogs, although the HR did not reliably decrease as might be expected. Dogs that showed normalization of BP within the first hour of fluid resuscitation were more likely to be discharged alive than those who remained hypotensive. A majority of dogs received and responded to isotonic crystalloids fluid resuscitation.

Comparison of regular insulin infusion doses in critically ill diabetic cats: 29 cases (1999-2007).

OBJECTIVE To compare biochemical parameters, neurologic changes, length of hospital stay, and clinical improvement in 3 groups of cats with diabetic ketosis/diabetic ketoacidosis (DK/DKA) prescribed varied doses of regular insulin as a continuous rate of infusion (CRI). DESIGN Retrospective study. SETTING University teaching hospital. ANIMALS Twenty-nine client-owned cats with DK/DKA prescribed a regular insulin CRI. INTERVENTIONS Cats were grouped as follows: 7 cats each in Group 1 and 2, (prescribed 1.1 and 2.2 U/kg/d, respectively), and 15 cats in Group 3 (prescribed increasing doses as needed). MEASUREMENTS AND MAIN RESULTS None of the groups received the total prescribed dose of insulin. The mean actual dose administered/kg/d ranged from 0.30 (0.21) to 0.87 (0.32) U/kg/d in Groups 1, 2, and 3. There was no difference in mean minimum blood glucose (BG) per 4 hours or change in BG from baseline per 4 hours between Groups 1 and 2 (P=0.63, 0.50). There was no difference between groups regarding the time required to reach a BG ≤ 13.9 mmol/L (250 mg/dL), serum phosphorus or potassium concentrations relative to baseline values (P=0.53, 0.90), length of time until urine or serum ketones were no longer detected (P=0.73), the animal commenced eating (P=0.24), or length of hospital stay (P=0.63). Four of the cats had declining mentation during hospitalization; there were no relationships between osmolality at presentation, either prescribed or administered insulin dose, and mentation changes. Three of the 4 cats with declining mentation survived. Twenty-seven of the 29 cats (93%) survived to discharge. CONCLUSIONS In this study, prescribing the published canine dose (2.2 U/kg/d) of regular insulin to cats with DK/DKA does not appear to increase the frequency of adverse neurologic or biochemical sequelae compared with cats that are prescribed the published cat dose (1.1 U/kg/d). The use of a sliding scale for determination of infusion rates significantly reduces the amount of insulin cats receive in this setting. Determination of whether adverse sequelae would occur more frequently if cats with DK/DKA received the full insulin prescribed doses of 1.1, 2.2, or >2.2 U/kg/d is warranted. Further controlled studies are necessary to determine if higher doses of insulin are associated with beneficial effects on morbidity or mortality.

Relationship between Doppler blood pressure and survival or response to treatment in critically ill cats: 83 cases (2003–2004)

OBJECTIVE To evaluate the relationship between Doppler blood pressure (DBP) and survival or response to treatment in critically ill cats. DESIGN Retrospective case series. ANIMALS 83 cats. PROCEDURES Medical records from cats admitted to the intensive care unit with at least 2 recorded DBP measurements were included in the study. Hypotension was defined as 1 or more DBP measurements d 90 mm Hg. Change in blood pressure, survival to hospital discharge, heart rate, rectal temperature, PCV, plasma pH, serum ionized calcium concentration, disease process, body weight, age, duration of hospitalization, and catecholamine treatment were also evaluated. RESULTS 39 cats were included in the hypotensive group, and 44 were consistently normotensive. Overall survival rate was 53% (44/83), with a significantly higher mortality rate in the hypotensive group (64% vs 32%). Among other variables, only low rectal temperature and low PCV were significantly associated with hypotension. Hypotensive cats with an increase in blood pressure of >or=20 mm Hg during hospitalization were more likely to survive to discharge (mortality rate, 69% vs 17%). CONCLUSIONS AND CLINICAL RELEVANCE Hypotensive cats had increased mortality rate with lower rectal temperatures and lower PCV, compared with normotensive critically ill cats. The implications of these findings with regard to treatment remain to be elucidated, but addressing these abnormalities may be appropriate.

Controversies in the use of fresh frozen plasma in critically ill small animal patients.

Objective To review the literature supporting or discouraging the use of fresh frozen plasma (FFP) transfusion in critically ill patients. Data Sources Human and animal publications were searched using PubMed without time limits and the following keywords were used: “fresh frozen plasma,” “coagulopathy,” “hypocoagulable state,” “hypercoagulable states,” and “critical illness.” Human Data Synthesis The commonly used tests of coagulation (eg, prothrombin time, activated partial thromboplastin time, international normalized ratio) are poorly predictive of clinical bleeding. FFP use in critically ill patients is unlikely to result in improved outcomes and may be associated with increased risks. Veterinary Data Synthesis There is insufficient evidence to make definitive conclusions regarding the use of FFP in critically ill animals, but clinical studies are underway that may provide further data that clarify the optimal use of FFP in animals. Conclusions The use of FFP in critically ill patients remains controversial. In the absence of clinical bleeding or a risk for clinical bleeding associated with a planned procedure, treatment use of FFP is not recommended in human patients. There are insufficient data in critically ill animals to enable formulation of recommendations. Further research is warranted in dogs and cats to establish evidence-based guidelines.

Survival analysis of critically ill dogs with hypotension with or without hyperlactatemia: 67 cases (2006–2011)

OBJECTIVE To determine whether critically ill hypotensive dogs without hyperlactatemia have the same prognosis as critically ill hypotensive dogs with hyperlactatemia. DESIGN Retrospective case series. ANIMALS 67 critically ill dogs with hypotension. PROCEDURES Medical records were searched from January 2006 through December 2011 for dogs that were hospitalized in the intensive care unit and that had hypotension and measurement of blood lactate concentration. Blood lactate concentration, systolic blood pressure, and survival rate were compared between hypotensive dogs with and without hyperlactatemia. RESULTS 19 of 67 (28%) dogs survived and were discharged from the hospital. Hypotensive dogs without hyperlactatemia had a significantly higher systolic blood pressure and were 3.23 (95% confidence interval, 1.04 to 9.43) times as likely to survive, compared with hypotensive dogs with hyperlactatemia. Age, weight, severity of clinical illness, and duration of hospitalization did not differ significantly between hypotensive dogs with and without hyperlactatemia. CONCLUSIONS AND CLINICAL RELEVANCE Results suggested that hypotensive dogs without hyperlactatemia had a better prognosis and chance of surviving to hospital discharge than did hypotensive dogs with hyperlactatemia. Because blood lactate concentration was negatively associated with systolic blood pressure and survival probability, it may be a useful metric for determining the prognosis of hypotensive dogs.

The association between vascular endothelial growth factor levels and clinically evident peripheral edema in dogs with systemic inflammatory response syndrome

OBJECTIVE To determine the relationship between plasma vascular endothelial growth factor (VEGF) levels, severity of illness, and edema formation in critically ill dogs. DESIGN Prospective, observational, descriptive, clinical study. SETTING University Teaching Hospital. ANIMALS Twenty-eight dogs. INTERVENTIONS None. MEASUREMENTS AND MAIN RESULTS Physical examination and multifrequency bioimpedance (MFBIA) measurements were performed daily on 28 critically ill dogs with evidence of severe inflammatory disease and compared with the corresponding plasma VEGF levels. The change in VEGF values and the relationship between MFBIA measurements and clinical evidence of edema were also examined. Eighteen dogs had a positive VEGF level, 12 dogs had clinical evidence of edema, and 7 dogs had both a positive VEGF level and clinical evidence of edema. There was no statistically significant correlation between VEGF levels and the presence of edema on physical examination (P=0.2). VEGF values were also evaluated with respect to WBC count, survival prediction index, presence of known sepsis, change in extracellular water, and outcome. No statistical relationship could be identified between VEGF levels in the blood of dogs with inflammatory disease and their survival prediction index (P=0.1), the WBC count (P=0.2), or presence of sepsis (P=0.2). Dogs with a VEGF level >70 pg/mL (n=3) were less likely to survive (P=0.04). Because of high variability within and between animals, conclusions regarding changes in MFBIA could not be made, suggesting that this technology requires further refinement and investigation in critically ill dogs. CONCLUSIONS A relationship between VEGF and clinically evident increased vascular permeability was not found in this study. Dogs with markedly elevated VEGF levels may be more likely to die, but further studies are needed to determine the diagnostic and prognostic value of VEGF in critically ill dogs.

Measurements of microvascular perfusion in healthy anesthetized dogs using orthogonal polarization spectral imaging.

OBJECTIVE To determine normal microvascular assessment parameters for healthy, anesthetized dogs. DESIGN Prospective investigational descriptive study. SETTING University Teaching Hospital. ANIMALS Fifteen client-owned, systemically healthy dogs that were undergoing general anesthesia for an elective procedure. INTERVENTIONS A sidestream dark-field videomicroscope probe was placed in the mouth at the mucogingival junction above the canine tooth and 3 video recordings of the microcirculation were made for later analysis by 2 independent, blinded reviewers. MEASUREMENTS AND MAIN RESULTS The videos were analyzed to determine the total vessel density, proportion of perfused vessels, microcirculatory flow index, and perfused vessel density. A range of values for these indices were obtained and reported. CONCLUSIONS The microcirculation of normal dogs is readily observable using the videomicroscope and recorded video segments can be used to determine microcirculatory measurements. These values may prove useful for comparison in future studies that examine canine microcirculatory parameters.

Controversies regarding choice of vasopressor therapy for management of septic shock in animals.

Objective To review and appraise common vasopressor drugs used to treat septic shock-induced hypotension in volume replete animals. Data Sources Human and animal publications were searched using PubMed without time limits and the following keywords were used: “vasopressor,” “septic shock,” “norepinephrine,” “dopamine,” “epinephrine,” and “vasopressin.” Human Data Synthesis The choice of vasopressor drug is unlikely to have a marked impact on outcome, but the incidence of adverse events (eg, tachycardia) varies greatly between the various treatment options. In agreement with the 2012 Cochrane Database consensus, norepinephrine is the first-choice vasopressor to maintain a mean arterial pressure ≥65 mm Hg. If an additional agent is required, epinephrine should be administered. Low-dose vasopressin can be added to norepinephrine to either increase the arterial blood pressure to the target goal value or decrease the norepinephrine dose, but should not be used as the initial vasopressor. Dopamine is not recommended except in highly selected circumstances. Veterinary Data Synthesis There is insufficient evidence to make definitive conclusions regarding the treatment of naturally occurring septic shock, but clinical studies are underway to provide further data. Conclusions The treatment of hypotension in people or animals with septic shock is challenging and vasopressor therapy is associated with a variety of adverse effects. Further research is warranted in dogs and cats to establish evidence-based guidelines.