A Journal of Veterinary Emergency and Critical Care Papers Episode

*** If you find these podcasts useful and interesting, PLEASE click on the iTunes icon below, then the "View in iTunes" blue link, and then rate and/or review the podcast. That will be really great and much appreciated. Thank you! ***

Listen to Stitcher

Episode based on May/June 2016 issue of the Journal of Veterinary Emergency and Critical Care

Please get in touch to request a copy of any of the papers discussed in this episode so you can read and critique the paper yourself!

Balakrishnan A, Drobatz KJ, Reineke EL. Development of anemia, phlebotomy practices, and blood transfusion requirements in 45 critically ill cats (2009–2011). J Vet Emerg Crit Care 2016. 26(3):406-411.

Anaemia is a relatively common clinical finding in critically ill patients
Likely multifactorial
Repeated phlebotomy to collect blood samples for diagnostic testing may be one of the causes or at least one of the contributing factors. Has been demonstrated in people, especially children, and suggested anecdotally in small animal patients.

“Critically ill cats that develop significant anaemia are often treated with blood transfusions. Red blood cell transfusions can help improve oxygen carrying capacity and may improve survival. However, at least in people, blood transfusions are associated with an increased expense, longer hospital stays, and carry the risk of several potentially life-threatening medical complications such as immunological transfusion reactions, infectious diseases, transfusion associated circulatory overload, and transfusion related acute lung injury.”

Information on this topic is limited in the veterinary literature and the primary objectives of this study were to:

  • Describe the incidence and development of anaemia in critically ill cats
  • Document phlebotomy practices and transfusion requirements in these cats
  • Evaluate the association between these factors on both duration of hospitalisation and outcome

Retrospective study
University of Pennsylvania ICU from 2009 to 2011
Exclusion criteria:

  • Did not stay in ICU for more than 48 hours (to exclude stable post-operative cases recovering in ICU)
  • Documented to have anaemia secondary to underlying chronic kidney disease
  • Incomplete medical records

Final sample size of 45 cats
Variety of primary diagnoses; respiratory disease, congestive heart failure and neoplasia most common
All of the cats were admitted via the emergency room in which they mostly stayed for less than 12 hours but some up to 24 hours
Approximately 20% were anaemic on admission to the hospital
40% developed anaemia before admission to the ICU (contribution of haemodilution from fluid therapy?)
Approximately 75% of the cats who were not anaemic on ICU admission went on to develop anaemia while in the ICU

Cats that developed anaemia after admission to the ICU had a significantly longer duration of hospitalisation than cats that did not develop anaemia while in the ICU.
Median duration: anaemic group 5 days, non-anaemic group 4 days
Range: anaemic group 3-24 days, non-anaemic group 3-13 days
Development of anaemia in the ICU was not statistically associated with outcome, just duration of hospitalisation.

Cats that required a blood transfusion for anaemia were found to have a significantly longer duration of hospitalisation but transfusion was not statistically associated with outcome. 

Phlebotomy:

  • Median number of phlebotomies per day for all cats in the ICU was 3 (range 1–6).
  • The 20 cats that developed anaemia during their ICU stay had a significantly greater number of phlebotomies per day (median 3, range 1–5) than the 7 cats that did not develop anaemia (median 1, range 1–2).
  • Cats that required a pRBC transfusion had a significantly greater number of daily phlebotomies (median 3, range 1–6) than cats that did not require a transfusion (median 2, range 1–4).
  • Cats that had a sampling or central venous catheter had a significantly greater number of phlebotomies (median 3, range 1–6) than cats without either of these catheters (median 1, range 1–2).

Results suggest that cats that developed anaemia after admission to their ICU had a longer duration of hospitalisation, likewise cats that received a blood transfusion. And probably this is because these cats were sicker.
These cats also had more blood samples taken, again probably because they were sicker.
Vicious circle: sicker cats have more blood samples taken = more blood loss = tendency towards or worsening of anaemia
Sicker cats also more likely to have anaemia as a result of other factors, potentially including a poorer regenerative response

Authors did not attempt to calculate any sort of illness severity scores because of the limitations in getting all the necessary data.

“In light of our study findings, adoption of blood conservation strategies should be considered. Blood conservation strategies are widely advocated in human intensive care medicine, particularly in critically ill children and include minimizing daily routine diagnostic phlebotomies, use of small volume or pediatric phlebotomy tubes, point of care and bedside microanalysis, minimization of blood sample wastage, lowering transfusion thresholds and transfusing only in response to physiologic need, and removing central venous and arterial catheters when no longer needed for patient monitoring purposes.”

For cats that do not have an in-dwelling sampling catheter in place, venepuncture is not entirely benign or risk free; unnecessary sampling can also contribute to patient stress, distress and reduced welfare.

Beer KS, Drobatz KJ. Severe anemia in cats with urethral obstruction: 17 cases (2002–2011). J Vet Emerg Crit Care 2016. 26(3):393-397.

“We hypothesized that cats with urethral obstruction and severe anemia requiring transfusion would have higher morbidity and mortality than cats with urethral obstruction without severe anemia.”

From an evidence-based perspective, this study was not able to prove or disprove this hypothesis.

Retrospective study from University of Pennsylvania over nine year period
Several limitations with respect to materials and methods, including small sample size:

  • 46 tomcats with urethral obstruction and anaemia
  • 17 tomcats met inclusion criteria, one of which was a PCV during hospitalisation of less than or equal to 20%
  • 2132 tomcats were treated for urethral obstruction during study period; 17 study cats = incidence of 0.8%

Authors suggest severe anaemia may largely be due to haemorrhage into urinary bladder – but this study does not provide evidence for this suggestion.

Full AM, Barnes Heller HL, Mercier M. Prevalence, clinical presentation, prognosis, and outcome of 17 dogs with spinal shock and acute thoracolumbar spinal cord disease. J Vet Emerg Crit Care 2016. 26(3): 412–418.

“Spinal shock is uncommonly reported in veterinary medicine and occurs when the spinal reflex arcs are anatomically normal but the patient exhibits transient hyporeflexia or areflexia caudal to a lesion….This is followed by a period of gradual return of the segmental spinal reflexes, and eventually hyperreflexia days to months later…In dogs with spinal shock the neurologic examination may yield a multifocal disease process or a lesion within the reflex arc, which could lead a clinician to an inaccurate neuroanatomic localization and differential diagnoses, and inappropriate diagnostic and treatment plan. An increased awareness of the prevalence, clinical presentation, common etiologies, and progress of spinal shock will aid the clinician in recognizing this syndrome.”

Upper motor neuron (UMN) lesion: expect hyperreflexia
Lower motor neuron (LMN) lesion: expect hyporeflexia
Authors key point is: If you examined a patient and found hyporeflexia you may suspect a LMN spinal reflex arc lesion when in fact the actual lesion is an UMN spinal cord lesion cranial to the localisation and the hyporeflexia is the result of spinal shock.

“The purpose of this study was to describe the prevalence and clinical presentation for dogs with thoracic vertebrae 3 (T3) to lumbar vertebrae 3 (L3) spinal lesions and suspected spinal shock.”

Retrospective study; November 2005 to 2010; private referral hospital in North America
986 dogs had spinal MRI performed
263 dogs remained after exclusion criteria applied
17/263 (6%) were diagnosed with spinal shock
94% of these 17 dogs presented within 24 hours of the onset of clinical signs 

Spinal shock following spinal cord injury has previously been described in association with severe spinal cord injury or transection causing loss of motor and sensory function in humans.
Also been observed and reported in a limited number of dogs with severe paraparesis or paraplegia.

“Our study is the first report specifically evaluating the prevalence and clinical presentation of spinal shock in dogs with acute thoracolumbar spinal injury.”

Spinal shock pathophysiology:

  • Studied in both human medicine and limited experimental veterinary studies
  • Complex syndrome
  • Underlying disease processes associated with spinal shock not been clearly defined

In people, a 4 phase model has described the alterations in spinal reflexes and time frame expected for return to function:

  • Phase 1: occurs within 0-24 hours; characterised by areflexia or hyporeflexia caudal to the spinal cord injury 
  • Phase 2: begins 1-3 days after injury; correlated with denervation hypersensitivity
  • Phase 3: 4-30 days post-injury; characterized by reappearance of deep tendon reflexes and the flexor withdrawal reflex
  • Phase 4: 1–12 months post-injury with return of all reflexes; reflexes often exaggerated during phase 4
“Mechanisms for recovery of spinal shock have been described including unveiling of latent synapses, alterations to the density or distribution of neurotransmitters and collateral sprouting of intact axons….The timing of segmental spinal reflex return has been suggested to be dependent on the individual’s amount and type of physical fitness prior to the injury. For example, highly trained athletes may have a shorter recovery of reflexes due to decreased tendon excitability, when compared to an untrained person.”

In this study, fibrocartilaginous embolism FCE) was most commons cause of spinal injury (7/17 dogs)
Acute non-compressive nucleus pulposus extrusion and intervertebral disk herniation were other causes

Results here indicate that dogs with clinical evidence of spinal shock have a high probability of at least partial neurological improvement:

  • 88% of dogs with documented neurological examinations at the time of discharge (1–12 days following diagnosis) had improved or normal reflexes, 75% of which specifically had improved withdrawal reflexes.
  • Remaining dogs lacking recorded neurological examinations at discharge, had improved or normal reflexes on subsequent recheck examinations with the exception of 1 dog.
  • Findings consistent with the previous literature suggesting reflexes often recover faster in non-primates compared to people
  • However, recovery of the withdrawal reflex was longer than 48 hours in many of the dogs in this study
“The lack of standardized follow-up time, especially in the immediate post-injury period, limits interpretation of the recovery process. A concise timeline of recovery is difficult in a retrospective study; therefore, caution should be taken when providing expected recovery times to clients.”
“In conclusion, although uncommon, spinal shock should be considered in any dog presenting with an acute history of thoracolumbar spinal injury and reduced reflexes in the pelvic limbs. Imaging should be pursued between the T3-S3 spinal segments in these patients to account for lesions in the T3-L3 spinal cord segment, which may result in spinal shock. The presence of spinal shock should not dissuade a veterinarian from pursuing appropriate diagnostic testing and therapy for the underlying etiology.”

Swann JW, Maunder CL, Roberts E, et al. Prevalence and risk factors for development of hemorrhagic gastro-intestinal disease in veterinary intensive care units in the United Kingdom. J Vet Emerg Crit Care 2016. 26(3): 419–427.

In human medicine, stress-related mucosal disease (SRMD) refers to the development of erosive lesions of the stomach and intestines in patients admitted to intensive care units (ICUs) for management of severe illness.
SRMD covers a spectrum of disease, from superficial mucosal injury detectable only by gastroduodenoscopy to severe ulceration that results in clinically important haemorrhage.
Overt clinical bleeding due to SRMD was reported to occur in approximately 4% of human patients admitted to a group of ICUs in Canada…
…and development of this disease significantly increased the risk of death during the period of hospitalisation.

“Impaired perfusion of the gastric mucosal barrier (GMB) is the proximate cause of SRMD, but development of the disease is reflective of systemic changes in hemodynamic status and inflammatory cascade”
“several factors have been identified in human patients that increase the risk of development of SRMD…particularly respiratory failure necessitating mechanical ventilation and coagulopathy. Administration of prophylactic gastro-protectant medications reduces the risk of SRMD…but this may be associated with development of other complications, such as aspiration pneumonia, because increased gastric pH permits bacterial colonization of the stomach.”

Haemorrhagic gastro-intestinal (GI) disease has not been described specifically in veterinary ICUs

“The primary aim of this study was to determine the proportion of animals that developed overt hemorrhagic GI disease in veterinary ICU patients. It was hypothesized that this would occur at similar rates to those reported in human ICUs, and that dogs would develop the disease more frequently than cats based on previous evidence suggesting that the GI tract is not the shock organ of cats. Secondary aims were to investigate risk factors for the development of hemorrhagic GI disease, and to determine whether development of these signs was associated with mortality during the period of hospitalization.”

Retrospective multicentre study in three UK teaching hospital ICUs; a lot of the data was collected prospectively
All cases presenting consecutively to the ICUs were considered eligible for enrolment during the period of the study if they were hospitalised for at least 24 hours
Exclusion criteria:

  • History of haemorrhagic GI disease in 48 hours prior to hospitalisation
  • Developed signs of haemorrhagic GI disease within the first 24 hours after admission
  • Surgical procedures involving the GI or upper respiratory tracts
  • Presented with or developed epistaxis or haemoptysis
  • Presented for management of GI disease
  • Sustained 1 or more skull fractures

Cases were not excluded if they:

  • Received gastro-protectant drugs, NSAIDs, glucocorticoids, or anticoagulants prior to admission or during hospitalisation
  • Were diagnosed with diseases that may cause secondary GI signs, such as hypoadrenocorticism

SRMD was defined as haemorrhagic GI disease manifesting as hematemesis, melena, or haematochezia or as mucosal erosions and haemorrhage observed during GI endoscopy.

Final sample size: 272 dogs and 94 cats
Some results:

  • 7.0% (CI: 4.5–10.7) (= 19 dogs) of dogs and no cats across the three centres developed SRMD
  • Among the dogs that received prophylactic gastro-protectant medications, the proportion that developed SRMD was 16.4% (CI: 8.9–28.3), compared to only 4.2% (CI: 2.2–7.8) in dogs that did not receive prophylaxis
  • Decreased serum albumin concentration, the ICU in question, and administration of prophylactic gastro-protectant medications were risk factors for the development of SRMD.
  • The proportion of dogs with SRMD that did not survive to discharge was significantly greater than for dogs that did not develop SRMD
  • Placement of a feeding tube and development of SRMD were associated with mortality

SJ comment:

“Now look as always, please don’t just take these points at face value and start repeating them. That would be entirely inappropriate. For starters we would need more studies, ideally prospective and blinded where possible, to evaluate all of this and demonstrate repeatability. And even then we would need to still be careful to distinguish association from causation.”

 Authors:

“Limitations of this study include the relatively small number of cases included, especially for investigation of risk factors for development of SRMD and mortality…it is possible that unmeasured differences between centers could have acted as confounding or modifying factors. Although much of the data included in this study were collected prospectively, some information regarding development of GI disease was collected retrospectively from clinical records, reducing the reliability and consistency of these findings. Data were also collected by a number of different investigators who may not have been involved in the primary care of the case.

Conclusions:

SRMD was observed in dogs from 3 different veterinary ICUs but was not observed in cats. Decreased serum albumin concentration was associated with development of SRMD, but, using a clinically relevant cut off value, this variable had a poor sensitivity and specificity for prediction of the disease. Development of SRMD and placement of a feeding tube were independently associated with increased mortality while hospitalized, but further studies will be required to determine the effects and potential benefits of prophylactic gastro-protectant therapy in veterinary ICU patients.”

If you would look a copy of any of the papers mentioned in this episode, let me know.

[This podcast is closely aligned with the MedEdLIFE Research Collaborative's Quality Checklist for Podcasts.]

Tweet: Check out FREE audio podcasts from @VetEmCC http://ctt.ec/UqL8b+ Also available in iTunes/Stitcher. #veterinary #podcast