Discussion:
To the authors’ knowledge, this report details the successful management of hypoxemic respiratory failure in a dog with complicated babesiosis. While definitive evidence of ARDS was not obtained with histopathology, it remains a likely differential diagnosis. Reports of management of suspected ARDS are scarce in the veterinary literature, and this report provides clinical guidance for such patients.
Acute respiratory distress syndrome (ARDS) describes the complex clinical syndrome of lung injury leading to downstream consequences of activation of cell-mediated immune pathways, endothelial injury, and alveolar flooding with protein-rich edema.10,11 This initial “exudative” phase of lung injury can be initiated secondary to several pulmonary and extra-pulmonary risk factors, but ultimately leads to increased vascular permeability, pulmonary edema, atelectasis, impaired gas exchange/hypoxemia, and respiratory fatigue often requiring positive pressure ventilation. Consensus definitions of acute lung injury and ARDS have been proposed previously for small animals.12 Based on these criteria, the authors propose that the dog described here fulfilled VetARDS criteria based on the following: (1) acute onset (<72 hours) tachypnea, (2) known risk factors (infection/sepsis, multiple transfusions), (3) and pulmonary capillary leak as evidenced by bilateral lung infiltrates on thoracic radiographs, proteinaceous fluid within the conducting airways, and no echocardiographic evidence supporting cardiogenic pulmonary edema. The fourth criterion, evidence of inefficient gas exchange, was not directly documented at the time of respiratory decline but arterial blood gases sampled after 12 hours of MV showed a profound hypoxemia and coupled with the findings of poor oxygen saturation immediately after intubation, prior to MV inefficient gas exchange was undoubtedly present.
The case presented here did have growth of oropharyngeal contaminant microbes on airway culture, but concurrent aspiration pneumonia cannot be completely ruled out. Several other factors make bacterial pneumonia less likely: the radiographic pulmonary changes were not classic for aspiration pneumonia, the patient presented afebrile, had normal lung sounds, and the tachypnea at presentation resolved after initial pRBC transfusion.
Respiratory failure in people with babesiosis that meet ARDS criteria is unique from other forms of ARDS because of its delayed onset and relatively rapid resolution when compared to more typical causes. One patient who survived severe hypoxemic respiratory failure with initiation of MV was extubated on the 10th day of treatment and discharged on day 20.4 Another describes a babesiosis-associated ARDS patient who was extubated on day 4 of hospitalization.6 The dog described here was on MV for approximately 2 days, with discharge on day 5 of hospitalization and developed suspect ARDS several weeks into his disease course. Respiratory compromise possibly attributable to ARDS has been documented in 6% of dogs with complicated babesiosis in one study, and only one dog was reported to recover, representing 3.2% (1/31) of complicated infections and 1.6% (1/63) of all infections.9 While the relative severity of respiratory compromise was not specified in the manuscript, all dogs were treated with oxygen supplementation.9 In one retrospective canine study of complicated babesiosis, 38% (32/84) of dogs died, of which 56% (18/32) had respiratory failure as the listed cause of death.4Another report of respiratory distress and mental obtundation in a dog with babesiosis had histopathology changes consistent with ARDS on necropsy.9 Interestingly, previous reports noted thatB. canis vogeli is associated with less severe disease except in the subpopulation of very young dogs,3 which fits the scenario presented here.
In people, ARDS historically carried a mortality rate of 40-50%, but trials in the last 20 years have shown improved outcomes with “lung-protective” ventilation strategies that include high PEEP and low tidal volumes to avoid alveolar overdistention and atelectrauma,13 prone positioning,14neuromuscular blockade,15 and early use of corticosteroids16 in certain subgroups of patients. There are no studies specifically evaluating ventilation strategies of ARDS and their impact on outcome in dogs. The dog in the present report was managed with ventilation settings adapted from the human ARDSnet trial.13 In this case, use of PEEP titrated to FiO2 requirements combined with neuromuscular blockade resulted in rapidly improved hypoxemia and pulmonary compliance. For the majority of the patient’s MV, a slightly higher tidal volume (55 mL, ~9.6 mL/kg) was used than the lung-protective tidal volumes from ARDSnet. It is unknown whether the same tidal volumes used in people with ARDS should be applied to dogs, as the proportion of lung:body weight may differ. The tidal volume in this patient was not reduced further due to hypercapnia.
At the time of initiation of neuromuscular blockade, the dog was considered heavily sedated. It is standard of care in people to meet certain criteria of sedation prior to using paralytic agents for MV, such as the Richmond agitation-sedation score [RASS].17 Ensuring a RASS of -4 to -5 (deep or unarousable sedation, respectively) helps avoid inadvertent awake paralysis in ventilated patients, which may cause ICU delirium and post-traumatic stress disorder from intensive care.16While human sedation criteria applied to dogs have not been evaluated, we would have scored the dog a RASS of -5 despite being tachypneic/dyssynchronous with the ventilator. The amount of sedation provided was not changed while the dog was paralyzed to avoid awake paralysis, but an optimal strategy to determine proper sedation prior to neuromuscular blockade remains unclear for dogs.
The present case report has several important limitations. Confirmation of ARDS with histopathology and full echocardiography by a cardiologist were not performed. Given that the dog responded with the therapy described and tolerated IV fluid administration, the authors feel that cardiovascular fluid overload is unlikely. Transfusion-related acute lung injury cannot be ruled out, though a blood type and crossmatch were performed prior to transfusion. In people this complication is more commonly associated with products containing soluble antibodies such as fresh frozen plasma, but activation of primed resident pulmonary leukocytes by other blood products secondary to severe systemic inflammation is another proposed mechanism.18 While we suspect that respiratory failure was already developing prior to blood transfusion, this cannot be proven. Regardless, receiving multiple transfusions is listed as a risk factor for ARDS in people as well as the veterinary consensus definition.12