Audit of the provision of nutritional support to mechanically ventilated dogs and cats

Objectives: To evaluate the use of enteral and parenteral nutrition in a population of mechanicallyventilatedcatsanddogs,identifyfactorsassociatedwithimplementation of nutrition, and assess the frequency of nutritional support within 72 hours of absent caloric intake. Design: Retrospective, single-center audit from June 2013 to June 2016. Setting: ICU of a veterinary university teaching hospital. Animals: Fifty-eight animals (50 dogs, 8 cats) that underwent mechanical ventilation for ≥ 6 hours with complete medical records. Interventions: None. Measurements and Main Results: Data collected included nutritional provision, time to initiation of nutrition, period of absent caloric intake, percentage of caloric intake obtained, and possible factors contributing to the delay or failure to implement nutrition. Thirty-one percent of patients (dogs 16/50, 32%; cats 2/8, 25%) received nutritional support during mechanical ventilation with all but 2 dogs receiving parenteral nutrition. Of those patients that did not receive nutrition (dogs 34/50, 68%; cats 6/8, 75%), documented contraindications or notations within the medical record for its omission were present in 16 of 34 dogs (47%) and 4 of 6 cats (66.7%). Thirteen animals (11 dogs, 2 cats)


INTRODUCTION
Clinical audits can be a valuable tool for assessing provided care against assumed or published standards and guidelines. 1 Audits of topics relating to nutrition are common in people and can drive improvements in clinical practice. [2][3][4][5][6][7] At present, there are no specific guidelines for nutritional support in dogs and cats undergoing mechanical ventilation (MV), although for general small animal populations several sources recommend the provision of nutritional support when the daily resting energy requirement (RER) has not been achieved for a period of 3 or more days. [8][9][10] In people with severe respiratory disease, poor nutritional state causes decreased respiratory drive to hypoxemia, 11 reduced surfactant production, 12 respiratory muscle catabolism and weakness resulting in difficulty weaning from MV, 13 and altered respiratory defense mechanisms. 14 Excessive nutrition is also harmful, with excessive carbohydrate administration resulting in hypercarbia and prolonged dependence on MV in critically ill people. 15,16 The appropriate provision of nutritional support in critically ill ventilated people is proven to reduce morbidity, mortality, and length of hospitalization, and to increase ventilator-free days. [17][18][19][20][21] Timing and modality of nutritional support are also important: in people, enteral nutrition (EN) is preferred and instituting EN within 48 hours of MV has been documented to reduce morbidity and mortality in numerous studies and metaanalyses. 13,20,22 To the authors' knowledge, there are no studies evaluating the provision of nutrition to critically ill mechanically ventilated cats and dogs; thus, the aim of this investigation was to perform a clinical audit of this population at our institution. Additional aims included to elucidate factors that may contribute to absent or delayed nutrition and to assess the rate of implementation of nutritional support within 72 hours of known absent caloric intake.

Patient population
Electronic patient records at the Queen Mother Hospital for Animals,

Statistical analysis
Data distribution was assessed for normality using the Shapiro-Wilk test. Normally distributed data were reported as mean and standard deviation (SD) and nonnormally distributed data as median and range.

RESULTS
Eighty-six patients were identified for possible inclusion into the audit (78 dogs, 8 cats). Eighteen dogs were excluded because insufficient records were available for appropriate data collection and 10 dogs were excluded because the duration of MV was <6 hours. Following exclusions, 50 dogs and 8 cats were included in the audit population ( Figure 1).

Dogs
Indications for initiating MV were hypoxemia (28/50, 56%), hypoventilation (7/50, 14%), respiratory fatigue (5/ there was no obvious contraindication to either modality of nutritional provision and no notation in the medical record outlining why it had been withheld.

Complications
Seven  Of the remaining 5 dogs, only 1 had a documented contraindication noted in the medical record (ie, marked refractory ileus) to feeding via the device. It is common for animals weaned from MV to require ongoing oxygen supplementation and occupying a nostril with an enteral device may increase airway resistance or limit the maximum achievable FiO 2 with nasal supplementation. Although an esophagostomy tube with an intraluminally placed gastric, duodenal, or jejunal tube would avoid this problem, this specific procedure is cost prohibitive at the authors' institution. The cost of placing such a device could reduce finances available for the patient to remain on MV. These factors may account for the very low level of enteral feeding devices seen in this audit.
Due to differences between human and veterinary ICU populations, caution is warranted in the direct translation of human therapeutic guidelines to animals. What seems rational to conclude from the current human and veterinary data is that early provision of nutritional support is beneficial, even when caloric requirements are not fully met.
Indeed, in mechanically ventilated people randomized to receive either hypocaloric "trophic" feeding or full caloric provision enterally for the first 6 days of MV, there was no difference in mortality or ventilatorfree days, but a reduced incidence of gastrointestinal intolerance to feeding in the trophic group. 25 The most obvious limitation of this audit is that it outlines nutritional support at a single institution, which therefore limits its applicability to other centers. Despite this limitation, it is the first such audit of nutrition in cats and dogs receiving MV and is intended to highlight barriers to nutritional support in this population. Other limitations include the inherent retrospective nature of audits and include reliance on appropriate data recording in the patients' medical records. Although records that lacked the objective data were excluded from the audit, the assessment of contributory factors for delayed or absent nutrition was hampered by the lack of notes in the medical record. It also became apparent that specific contraindications to nutrition could have been present but not noted in the medical record. In cases where the exact length of time with no caloric intake prior to hospitalization could not be precisely determined, we opted to use the nearest 12-hour period of known absent caloric intake. This will have resulted in some patients having absent intake for longer than calculated and thus appear to be at reduced need for nutritional provision. Similarly, the prehospitalization period of absent caloric intake was unknown for 6 dogs and 4 cats, which likely led to underestimation of nutritional requirements in these cases. The difficulty in being able to quantify nutritional status prior to hospital admission, along with the very low number of patients with a recorded BCS, made rigorous assessment of the need for nutritional provision prior to ventilation difficult. Although no mechanical or septic complications were noted in patients receiving PN in this audit, it is possible that some animals suffered such complications but that these were not noted within the medical record. As previously mentioned, specific institutional protocols and costs (such as the need to reserve a vascular port for PN at catheter placement and high cost to place gastrostomy and duodenostomy tubes) will further limit the usefulness of this audit to institutions other than our own. Despite its limitations, this audit highlights a lack of adequate nutritional provision for many patients at this institution, with many having no obvious reason for this oversight.
The majority of animals described in this clinical audit received NNS (68% of dogs, 75% of cats), and in many cases a well-defined reason could not be determined from the medical record. Although many animals in this audit had contraindications to one feeding modality, it remains unclear why another modality was not used. In light of various proposed clinical standards, clinicians should be cognizant of the potential benefits of nutrition in the critically ill, mechanically ventilated dog and cat. Clinical audits, especially when repeated after