Physiological effects of rapid intravenous infusion of fluids for the treatment of hypovolaemic hypotension in anaesthetised cats

Abstract

My research focused on providing information to further our understanding on the physiology of severe haemorrhage and the administration of fluids to resuscitate the intravascular compartment in anaesthetised domestic cats. To date, there are very few reports detailing evidence-based treatments on how to manage cats in these clinical situations. The current treatment recommendations are based on decades of anecdotal textbook-based evidence, or are recommendations extrapolated from dogs but adjusted to fit cat-related differences. Although current treatments are effective in many cats, the risk of complications is also very high. Fluid overload occurs commonly during fluid resuscitation. Therefore cats are frequently claimed to be a ‘fluid-sensitive’ species and caution is strongly advised when administering intravenous fluids during hypovolaemic resuscitation. This fluid-sensitive sentiment can easily lead to under-treatment or, if complications arise during management, attribution of blame to the fluids used as the most likely cause of treatment failure. These situations can delay or prevent a cat from receiving the correct treatment as soon as possible to prevent lingering shock, or can result in aggressive overtreatment and fluid overload and its sequalae. A comprehensive understanding of the cat’s physiology during severe haemorrhage and fluid resuscitation will demystify fluid treatment protocols and allow for clinically meaningful recommendations to be made. This thesis comprises of a series of investigations which focus on physiological responses to: 1) acute severe haemorrhage and 2) fluid administration to treat haemorrhage-induced hypovolaemic hypotension. Six domestic cats underwent three treatments, two month apart, in a randomised crossover pattern. The cats were anaesthetised using: 1) buprenorphine as a pre-anaesthetic medication, 2) alfaxalone for induction into general anaesthesia, and 3) isoflurane in oxygen for maintenance of general anaesthesia. The three treatments were: 1) control, with sham haemorrhage and resuscitation; 2) controlled haemorrhage followed by lactated Ringer’s solution for resuscitation (LRS); and 3) controlled haemorrhage followed by tetrastarch 6% 130/0.4 for resuscitation. Cats that underwent a controlled haemorrhage phase, followed by a resuscitation phase were administered LRS or tetrastarch 6% 130/0.4 at 60- and 20-ml kg-1 per hour, respectively, for 120 minutes. The endpoint of haemorrhage was a 1) maximum of 30 ml kg-1 of blood withdrawn, or 2) invasively measured mean

iv | P a g e

arterial blood pressure < 48 mmHg for 3 minutes. This controlled haemorrhage model in cats while under general anaesthesia emulates a clinically relevant scenario, whereby cats undergoing surgical procedures could haemorrhage and the surgeon could gain control of the haemorrhage. The first study in this thesis describes changes in physiological, haematological and biochemical variables following a mild and severe haemorrhage event. Variables that changed significantly were further analysed to determine if they could be used in a scoring system aimed to quantify acute haemorrhage during general anaesthesia. We found that easily determinable variables could be placed into four ratios that, when analysed together, can reliably quantify acute blood loss in anaesthetised cats. We called this scoring system the Cat Acute Bleeding Scoring System, or CABSS as an abbreviated term. The second study in this thesis describes changes in physiological, haematological and biochemical variable values during the fluid resuscitation phase where LRS or tetrastarch 6% 130/0.4 were administered. Bearing in mind that cats are regarded as fluid-sensitive, we specifically aimed to derive volume-endpoints to guide how much fluid should be administered to cats that have just undergone an acute severe haemorrhage. Volume-endpoints are unlike resuscitation endpoints which are commonly derived cardiovascular or oxygenation biomarkers used to guide fluid therapy in goal-directed fluid administration protocols. Resuscitation endpoints seek to restore normal physiological function, however, the lack of clear guidelines on how much fluid to administer before deciding a cat is a fluid non-responder predisposes them to fluid overload. During resuscitation, fluids are often required to fill the intravascular compartment, but there are no clear guide for veterinarians to stop administering fluids and rather begin other treatment modalities, such as administration of a catecholamine to restore and meet traditional resuscitation endpoint targets. My thesis is the first to describe such endpoints in cats that have undergone severe haemorrhage. We found that changes in the sodium, chloride and magnesium electrolyte concentrations were helpful in determining volume endpoints. Specifically, the chloride to sodium ratio appears to be the volume-endpoint of choice and is easy to measure and cost-effective, unlike many of the traditional resuscitation endpoints. The clinical goal is to administer resuscitation fluids until the volume-endpoint is reached; and if the cat still does not meet the desired resuscitation endpoint, then catecholamines or other cardio-active drugs should be considered before administering more fluids, to prevent fluid overload. The third study describes the acid-base status of the cats during haemorrhage and resuscitation. We made use of three methods to analyse the acid-base status, namely: 1) Henderson-Hasselbalch, 2) Stewart, and 3) semi-quantitative

v | P a g e

methods. We found that severe haemorrhage and large volume resuscitation did not alter the blood pH, and acidosis was only detected during anaesthesia. Despite the pH being similar among treatments, there were method-associated differences in variables used to interpret pH. However, the clinical relevance of the acidosis is not yet fully understood. The fourth study describes the haemostatic effects of severe haemorrhage followed by fluid resuscitation. We found that haemostatic derangements occurred during the resuscitation phase of the LRS and tetrastarch 6% 130/0.4 treatments, and we attributed these to haemodilution. The findings of these studies have clinical relevance and allow us to quantify acute intraoperative haemorrhage and guide fluid resuscitation, using two commonly available intravenous fluids for treatment of haemorrhage-induced hypovolaemia. In order to improve our management of cats during the peri-anaesthetic period, we require evidence-based treatment guides and protocols which are aimed at improving the overall outcome of treating hypovolaemia in cats. It is essential that the CABSS and volume-endpoints be scrutinised and evaluated on a larger and more diverse population of cats requiring treatment for other pathophysiological presentations besides during the peri-anaesthetic period. The evidence provided in this thesis improves our overall understanding of the physiology of severe haemorrhage followed by fluid resuscitation, especially during the peri-anaesthetic period.