What is the value of statistical testing of observational data?

Statistical analysis of medical data aims to reveal patterns that can aid in decision making for future cases and, hopefully, improve patient outcomes. Large and bias-free datasets, such as those produced in formal randomized clinical trials, are necessary to make such analyses as reliable as possible. For a host of reasons, randomized trials are, unfortunately, relatively uncommon in veterinary medicine and surgery, implying that less ideal datasets (mostly observational data) must form the basis for much of our decision making regarding treatment of individual patients under our care. In this review, we first describe the common shortcomings of many observational veterinary datasets when viewed in comparison with their optimal counterparts and highlight how the deficiencies can lead to unreliable conclusions. We illustrate how many of the interpretative problems associated with observational data, predominantly various forms of bias, are not solved, and may even be exacerbated, by statistical analysis. We emphasize the need to examine summary data and its derivation in detail without being lured into relying upon P values to draw conclusions and advocate for completely omitting statistical analysis of many observational datasets. Finally, we present some suggestions for alternative statistical methods, such as propensity scoring and Bayesian methods, which might help reduce the risk of drawing unwarranted, and overconfident, conclusions from imperfect data.

Extended durotomy to treat severe spinal cord injury after acute thoracolumbar disc herniation in dogs.

OBJECTIVE: To report recovery of ambulation of dogs treated with extended thoracolumbar durotomy for severe spinal cord injury caused by intervertebral disc herniation. STUDY DESIGN: Descriptive cohort. ANIMALS: Twenty-six consecutive paraplegic dogs presented with loss of deep pain sensation after acute thoracolumbar intervertebral disc herniation. METHODS: Each dog underwent routine diagnostic assessment and surgery for removal of extradural herniated intervertebral disc, followed by a four-vertebral body length durotomy centered on the herniated disc. Each dog was followed up until it was able to walk 10 steps without assistance or until 6 months after surgery. RESULTS: Sixteen of 26 dogs recovered to walk unaided (all but one also recovered fecal and urinary continence), and six dogs did not; four dogs were lost to follow-up. One dog was euthanized because of signs consistent with progressive myelomalacia. There was no evidence of detrimental effects of durotomy within the period of study. Using Bayesian analysis, we found a point estimate of successful outcome of 71% with 95% credible interval from 52% to 87%. CONCLUSION: Extended durotomy seemed to improve the outcome of dogs in our case series without increase in morbidity. CLINICAL SIGNIFICANCE: Extended durotomy appears safe and may improve the outcome of dogs with severe thoracolumbar mixed contusion and compressive injuries associated with acute intervertebral disc extrusion.

Therapeutic efficacy of microtube-embedded chondroitinase ABC in a canine clinical model of spinal cord injury.

See Moon and Bradbury (doi:10.1093/brain/awy067) for a scientific commentary on this article.Many hundreds of thousands of people around the world are living with the long-term consequences of spinal cord injury and they need effective new therapies. Laboratory research in experimental animals has identified a large number of potentially translatable interventions but transition to the clinic is not straightforward. Further evidence of efficacy in more clinically-relevant lesions is required to gain sufficient confidence to commence human clinical trials. Of the many therapeutic candidates currently available, intraspinally applied chondroitinase ABC has particularly well documented efficacy in experimental animals. In this study we measured the effects of this intervention in a double-blinded randomized controlled trial in a cohort of dogs with naturally-occurring severe chronic spinal cord injuries that model the condition in humans. First, we collected baseline data on a series of outcomes: forelimb-hindlimb coordination (the prespecified primary outcome measure), skin sensitivity along the back, somatosensory evoked and transcranial magnetic motor evoked potentials and cystometry in 60 dogs with thoracolumbar lesions. Dogs were then randomized 1:1 to receive intraspinal injections of heat-stabilized, lipid microtube-embedded chondroitinase ABC or sham injections consisting of needle puncture of the skin. Outcome data were measured at 1, 3 and 6 months after intervention; skin sensitivity was also measured 24 h after injection (or sham). Forelimb-hindlimb coordination was affected by neither time nor chondroitinase treatment alone but there was a significant interaction between these variables such that coordination between forelimb and hindlimb stepping improved during the 6-month follow-up period in the chondroitinase-treated animals by a mean of 23%, but did not change in controls. Three dogs (10%) in the chondroitinase group also recovered the ability to ambulate without assistance. Sensitivity of the dorsal skin increased at 24 h after intervention in both groups but subsequently decreased to normal levels. Cystometry identified a non-significant improvement of bladder compliance at 1 month in the chondroitinase-injected dogs but this did not persist. There were no overall differences between groups in detection of sensory evoked potentials. Our results strongly support a beneficial effect of intraspinal injection of chondroitinase ABC on spinal cord function in this highly clinically-relevant model of chronic severe spinal cord injury. There was no evidence of long-term adverse effects associated with this intervention. We therefore conclude that this study provides strong evidence in support of initiation of clinical trials of chondroitinase ABC in humans with chronic spinal cord injury.

When neuroscience met clinical pathology: partitioning experimental variation to aid data interpretation in neuroscience.

In animal experiments, neuroscientists typically assess the effectiveness of interventions by comparing the average response of groups of treated and untreated animals. While providing useful insights, focusing only on group effects risks overemphasis of small, statistically significant but physiologically unimportant, differences. Such differences can be created by analytical variability or physiological within-individual variation, especially if the number of animals in each group is small enough that one or two outlier values can have considerable impact on the summary measures for the group. Physicians face a similar dilemma when comparing two results from the same patient. To determine whether the change between two values reflects disease progression or known analytical and physiological variation, the magnitude of the difference between two results is compared to the reference change value. These values are generated by quantifying analytical and within-individual variation, and differences between two results from the same patient are considered clinically meaningful only if they exceed the combined effect of these two sources of 'noise'. In this article, we describe how the reference change interval can be applied within neuroscience. This form of analysis provides a measure of outcome at an individual level that complements traditional group-level comparisons, and therefore, introduction of this technique into neuroscience can enrich interpretation of experimental data. It can also safeguard against some of the possible misinterpretations that may occur during analysis of the small experimental groups that are common in neuroscience and, by illuminating analytical error, may aid in design of more efficient experimental methods.

Development of an International Canine Spinal Cord Injury observational registry: a collaborative data-sharing network to optimize translational studies of SCI.

STUDY DESIGN: Prospective cross-sectional cohort study. OBJECTIVES: The canine spontaneous model of spinal cord injury (SCI) is as an important pre-clinical platform as it recapitulates key facets of human injury in a naturally occurring context. The establishment of an observational canine SCI registry constitutes a key step in performing epidemiologic studies and assessing the impact of therapeutic strategies to enhance translational research. Further, accumulating information on dogs with SCI may contribute to current "big data" approaches to enhance understanding of the disease using heterogeneous multi-institutional, multi-species datasets from both pre-clinical and human studies. SETTING: Multiple veterinary academic institutions across the United States and Europe. METHODS: Common data elements recommended for experimental and human SCI studies were reviewed and adapted for use in a web-based registry, to which all dogs presenting to member veterinary tertiary care facilities were prospectively entered over ~1 year. RESULTS: Analysis of data accumulated during the first year of the registry suggests that 16% of dogs with SCI present with severe, sensorimotor-complete injury and that 15% of cases are seen by a tertiary care facility within 8 h of injury. Similar to the human SCI population, 34% were either overweight or obese. CONCLUSIONS: Severity of injury and timing of presentation suggests that neuroprotective studies using the canine clinical model could be conducted efficiently using a multi-institutional approach. Additionally, pet dogs with SCI experience similar comorbidities to people with SCI, in particular obesity, and could serve as an important model to evaluate the effects of this condition.

Shunt tube placement for amelioration of cerebrospinal fluid flow obstruction caused by spinal cord subarachnoid fibrosis in dogs.

OBJECTIVE: To describe a novel technique for ameliorating cerebrospinal fluid flow obstruction secondary to pia-arachnoid fibrosis in dogs and report outcome. STUDY DESIGN: Descriptive report and retrospective case series. ANIMALS: Dogs with cerebrospinal fluid (CSF) flow obstruction (n = 7). METHODS: Medical records were searched for dogs that had a subarachnoid-subarachnoid shunt placed for treatment of CSF flow obstruction. Data collected included age, sex, breed, clinical signs and duration of signs prior to examination, neurologic status and localization prior to surgery, pre-surgical diagnostics, surgical technique, histopathology, postoperative neurologic examination, time to discharge from hospital, and outcome. RESULTS: All dogs were diagnosed at surgery with a fibrotic adhesion between the arachnoid and pia mater. A subarachnoid shunting tube was implanted to allow CSF flow across the lesion site. Five dogs showed improvement of clinical signs, 3 of which showed complete recovery and 2 of which showed improvement without resolution of all clinical signs. Two dogs showed no change at 7 and 24 months postoperatively. CONCLUSION: Bridging a region of pia-arachnoid fibrosis with a tube placed in the subarachnoid space can ameliorate or prevent progression of associated clinical signs.

SINGLE-SHOT TURBO SPIN ECHO PULSE SEQUENCE FINDINGS IN DOGS WITH AND WITHOUT PROGRESSIVE MYELOMALACIA.

Progressive myelomalacia is an uncommon type of ischemic, hemorrhagic spinal cord infarction. Diagnosis can be difficult, but prompt recognition is important. We hypothesized that cerebrospinal fluid signal attenuation on magnetic resonance (MR) images would be more extensive in dogs that developed progressive myelomalacia vs. control dogs. A retrospective analytic cohort study was designed. Dogs were included if they presented for acute paraplegia and loss of deep pain perception and had undergone MR imaging using both sagittal single-shot turbo spin echo (SSTSE) and standard sagittal T2-weighted fast spin echo (T2W) pulse sequences. Dogs were divided into progressive myelomalacia and control groups for comparisons. All MR examinations were evaluated by three reviewers blinded to patient outcome. Length of cerebrospinal fluid attenuation was recorded as a ratio to the length of the L2 vertebral body in SSTSE and T2W sequences (CSF:L2SSTSE and CSF:L2T2 , respectively). Length of intramedullary spinal cord hyperintensity was recorded as a ratio to the length of the L2 vertebral body in T2W sequences. A total of 21 dogs were included (five in the progressive myelomalacia group and 16 in the control group). The mean CSF:L2SSTSE attenuation value was significantly higher in dogs that developed progressive myelomalacia (CSF:L2SSTSE = 10.7) compared to controls (CSF:L2SSTSE = 5.4; P = 0.015). A cut off ratio of attenuation >7.4 provided optimal differentiation between groups in this study. Findings supported the conclusion that dogs with CSF:L2SSTSE ≤ 7.4 are unlikely to develop progressive myelomalacia while dogs with CSF:L2SSTSE > 7.4 are indeterminate for progressive myelomalacia.

Arachidonic acid pathway alterations in cerebrospinal fluid of dogs with naturally occurring spinal cord injury.

BACKGROUND: Canine intervertebral disc πherniation causes a naturally-occurring spinal cord injury (SCI) that bears critical similarities to human SCI with respect to both injury pathomechanisms and treatment. As such, it has tremendous potential to enhance our understanding of injury biology and the preclinical evaluation of novel therapies. Currently, there is limited understanding of the role of arachidonic acid metabolites in canine SCI. RESULTS: The CSF concentrations of PLA2 and PGE2 were higher in SCI dogs compared to control dogs (p = 0.0370 and 0.0273, respectively), but CSF LCT4 concentration in SCI dogs was significantly lower than that in control dogs (p < 0.0001). Prostaglandin E2 concentration in the CSF was significantly and positively associated with increased severity of SCI at the time of sampling (p = 0.041) and recovery 42 days post-injury (p = 0.006), as measured by ordinal behavioral scores. CONCLUSION: Arachidonic acid metabolism is altered in dogs with SCI, and these data suggest that these AA metabolites reflect injury severity and recovery, paralleling data from other model systems.

Removal of a Presumed Peripheral Cerebral Cyst via Craniectomy in a Crested Pekin Duck ( Anas platyrhynchos f dom).

A juvenile, male crested pekin duck ( Anas platyrhynchos f dom) was presented for neurologic signs suggestive of cerebellar disease. Physical examination revealed microphthalmia, erratic head movements, and ataxia. Computed tomography scan of the head and neck regions revealed 2 full-thickness skull-bone defects within the caudal portion of the cranium. The cerebellum appeared to be ventrally compressed by a homogeneous, triangular, fluid-attenuating region (0-10 Hounsfield units). A craniectomy was performed, and a presumed peripheral cerebral cyst was removed with suction and gentle dissection. No postoperative complications occurred, and the patient showed clinical improvement for 5 months after surgery. However, after 5 months, the owners elected euthanasia because of poor prognosis after finding the duck minimally responsive in a water enclosure. At necropsy, a thin-walled, epithelial structure was present in meninges and was adhered to the skull at the presumed surgical site.

Suspected primary hematomyelia in 3 dogs.

Primary hematomyelia refers to hemorrhage occurring within the spinal cord without an identifiable etiology. Clinical signs, magnetic resonance imaging characteristics, and histopathological findings are described. Diagnosis was made through histological analysis and rule-outs for underlying factors. Following removal of the hematoma, neurologic deficits improved, although some residual deficits persisted.

Hématomyélie primaire suspectée chez 3 chiens. L'hématomyélie primaire fait référence à l'hémorragie qui se produit dans la moelle épinière sans une étiologie identifiable. Les signes cliniques, les caractéristiques de l'imagerie par résonance magnétique et les résultats de l'histopathologie sont décrits. Le diagnostic a été posé à l'aide d'une analyse histologique et de l'élimination des facteurs sous-jacents. Après l'enlèvement de l'hématome, le déficit neurologique s'est amélioré, même si des déficits résiduels ont persisté.(Traduit par Isabelle Vallières).

Causal Agent Investigation and Treatment of Dogs Diagnosed with Discospondylitis in a Brucella canis Endemic Region.

Discospondylitis is a well-recognized disease in dogs, but the relative prevalence of causal infectious agents and efficiency of relevant diagnostic tests are not well-established. Medical record review identified 117 dogs diagnosed with discospondylitis in our clinic over a 5-year period. In 32 dogs, discospondylitis was diagnosed as an incidental imaging finding; 24 of these dogs had concomitant neoplasia. A likely causal infection was identified in 45 of the remaining 85 dogs in which blood and urine cultures, serology for Brucella spp., and galactomannan fungal antigen testing were recommended. Ten dogs were diagnosed with Brucella canis, and ten were diagnosed with suspected fungal infection. Brucella suis serology was negative in all 35 dogs that were tested. Blood cultures were positive in 28 of 71 (39%) tested dogs, and urine culture was positive in 12 of 79 (15%). Cultures were positive from the lesion site of four of eight dogs that underwent surgery and one of the five dogs that underwent image-guided lesion sample collection. Subluxation secondary to discospondylitis was stabilized with metallic implants in four dogs. A similar proportion of known satisfactory treatment outcomes at last follow-up were recorded in dogs that had suspected fungal disease, other bacterial infections, or were Brucella-positive and in those dogs with imaging diagnosis only, although some individuals continued to receive anti-microbial agents or showed recurrent signs. These data support the value of blood culture in discospondylitis and suggest a relatively high prevalence of infection with Brucella spp. and suspected fungal infection.

Randomized Controlled Trial of Durotomy as an Adjunct to Routine Decompressive Surgery for Dogs With Severe Acute Spinal Cord Injury.

Although many interventions for acute spinal cord injury (SCI) appear promising in experimental models, translation directly from experimental animals to human patients is a large step that can be problematic. Acute SCI occurs frequently in companion dogs and may provide a model to ease translation. Recently, incision of the dura has been highlighted in both research animals and human patients as a means of reducing intraspinal pressure, with a view to improving perfusion of the injured tissue and enhancing functional recovery. Observational clinical data in humans and dogs support the notion that it may also improve functional outcome. Here, we report the results of a multi-center randomized controlled trial of durotomy as an adjunct to traditional decompressive surgery for treatment of severe thoracolumbar SCI caused by acute intervertebral disc herniation in dogs. Sample-size calculation was based on the proportion of dogs recovering ambulation improving from an expected 55% in the traditional surgery group to 70% in the durotomy group. Over a 3.5-year period, we enrolled 140 dogs, of which 128 had appropriate duration of follow-up. Overall, 65 (51%) dogs recovered ambulation. Recovery in the traditional decompression group was 35 of 62 (56%) dogs, and in the durotomy group 30 of 66 (45%) dogs, associated with an odds ratio of 0.643 (95% confidence interval: 0.320-1.292) and z-score of -1.24. This z-score indicates trial futility to reach the target 15% improvement over traditional surgery, and the trial was terminated at this stage. We conclude that durotomy is ineffective in improving functional outcome for severe acute thoracolumbar SCI in dogs. In the future, these data can be compared with similar data from clinical trials on duraplasty in human patients and will aid in determining the predictive validity of the "companion dog model" of acute SCI.

Autologous olfactory mucosal cell transplants in clinical spinal cord injury: a randomized double-blinded trial in a canine translational model.

This study was designed to determine whether an intervention proven effective in the laboratory to ameliorate the effects of experimental spinal cord injury could provide sufficient benefit to be of value to clinical cases. Intraspinal olfactory ensheathing cell transplantation improves locomotor outcome after spinal cord injury in 'proof of principle' experiments in rodents, suggesting the possibility of efficacy in human patients. However, laboratory animal spinal cord injury cannot accurately model the inherent heterogeneity of clinical patient cohorts, nor are all aspects of their spinal cord function readily amenable to objective evaluation. Here, we measured the effects of intraspinal transplantation of cells derived from olfactory mucosal cultures (containing a mean of ~50% olfactory ensheathing cells) in a population of spinal cord-injured companion dogs that accurately model many of the potential obstacles involved in transition from laboratory to clinic. Dogs with severe chronic thoracolumbar spinal cord injuries (equivalent to ASIA grade 'A' human patients at ~12 months after injury) were entered into a randomized double-blinded clinical trial in which they were allocated to receive either intraspinal autologous cells derived from olfactory mucosal cultures or injection of cell transport medium alone. Recipients of olfactory mucosal cell transplants gained significantly better fore-hind coordination than those dogs receiving cell transport medium alone. There were no significant differences in outcome between treatment groups in measures of long tract functionality. We conclude that intraspinal olfactory mucosal cell transplantation improves communication across the damaged region of the injured spinal cord, even in chronically injured individuals. However, we find no evidence for concomitant improvement in long tract function.

Circulating neutrophil activation in dogs with naturally occurring spinal cord injury secondary to intervertebral disk herniation.

OBJECTIVE: To investigate the time course of circulating neutrophil priming and activity in dogs with spinal cord injury secondary to intervertebral disk herniation that undergo decompressive surgery. ANIMALS: 9 dogs with spinal cord injury and 9 healthy dogs (controls). PROCEDURES: For dogs with spinal cord injury, blood samples were collected on the day of hospital admission and 3, 7, 30, and 90 days after injury and decompressive surgery. A single blood sample was collected from the control dogs. Flow cytometry analysis was performed on isolated neutrophils incubated with antibody against CD11b and nonfluorescent dihydrorhodamine 123, which was converted to fluorescent rhodamine 123 to measure oxidative burst activity. RESULTS: Expression of CD11b was increased in dogs with spinal cord injury 3 days after injury and decompressive surgery, relative to day 7 expression. Neutrophils expressed high oxidative burst activity both 3 and 7 days after injury and decompressive surgery, compared with activity in healthy dogs. CLINICAL RELEVANCE: For dogs with spinal cord injury, high CD11b expression 3 days after injury and decompressive surgery was consistent with findings for rodents with experimentally induced spinal cord injury. However, the high oxidative burst activity 3 and 7 days after injury and decompressive surgery was not consistent with data from other species, and additional studies on inflammatory events in dogs with naturally occurring spinal cord injury are needed.

Cerebrospinal fluid flow on time-spatial labeling inversion pulse images before and after treatment of congenital hydrocephalus in a dog.

A 3-month-old male cross-breed dog presented with signs of progressive diffuse brain disease. Noncommunicating congenital hydrocephalus concurrent with cervical syringomyelia was diagnosed on magnetic resonance images. On time-spatial labeling inversion pulse (Time-SLIP) images CSF flow through the mesencephalic aqueduct was poorly defined and there was flow into the syrinx across the craniocervical junction. After percutaneous ventricular drainage and ventriculoperitoneal shunting, CSF flow through the aqueduct was clearly detected and flow into the syrinx disappeared. In addition, CSF flow in the subarachnoid space at the pons and ventral aspect of the cervical subarachnoid space was restored. Signs of neurological dysfunction improved after ventriculoperitoneal shunting and the cerebral parenchyma was increased in thickness on 2-year follow-up computed tomography images. Patterns of CSF flow on Time-SLIP images before and after CSF drainage or ventriculoperitoneal shunting aid in clarifying disease pathogenesis and confirm effects of CSF drainage.

Brachial artery thrombosis in a dog causing monoparesis mimicking nerve sheath tumor.

There are few differential diagnoses for non-orthopedic thoracic limb lameness in adult dogs aside from nerve tumors and disk-associated nerve compression; this report introduces another etiology. A 9-year-old male castrated mixed dog presented with an episodic history of nonweight-bearing thoracic limb lameness. Additional clinical signs included an atrophied thoracic limb with cool paw pads and painful axillary region. Magnetic resonance imaging, computed tomography, ultrasound, and exploratory surgery confirmed a chronic thrombus of the right brachial artery. No underlying cause for the thrombus was identified. The dog has been successfully managed on long-term rivaroxaban and clopidogrel. Follow-up ultrasound of the thrombus suggested early remodeling.

Variation in biochemistry test results between annual wellness visits in apparently healthy Golden Retrievers.

BACKGROUND: Annual wellness testing is widely recommended for apparently healthy dogs, but there is little data to assist with distinguishing normal variation from clinically important changes. OBJECTIVES: To define variability in biochemistry analytes between annual wellness tests in healthy Golden Retrievers. ANIMALS: Four hundred thirty-four Golden Retrievers undergoing annual health assessments by their primary care veterinarians as part of a prospective cohort study. METHODS: Changes in 23 biochemistry analytes were calculated between year 1 and year 2 health checks for 196 dogs classified as healthy for ≥3 consecutive years. Using a direct nonparametric method, annual change intervals were constructed to define normal variability. A validation cohort of 238 dogs without a diagnosis of systemic disease for ≥3 consecutive years were compared with the reference and annual change intervals, and the proportions of dogs outside annual change intervals and a population-based reference interval were compared by using a McNemar test. RESULTS: Annual change intervals were calculated based on 190 dogs after outlier removal. For all 23 analytes, >90% of dogs in the validation cohort were within the annual change interval. There were no significant differences in the classification by reference versus annual change intervals. CONCLUSIONS AND CLINICAL IMPORTANCE: The annual change intervals met performance requirements for classification of dogs that did not develop systemic disease in the year following wellness testing as normal.

Association of magnetic resonance assessed disc degeneration and late clinical recurrence in dogs treated surgically for thoracolumbar intervertebral disc extrusions.

BACKGROUND: Radiographic signs of intervertebral disc mineralization are thought to indicate sites of future recurrence of disc extrusion (Hansen type I) but the relationship between evidence of disc degeneration on magnetic resonance imaging (MRI) and future disc extrusion with recurrence of clinical signs has not been examined. OBJECTIVES: To examine the relationship between MRI-assessed degeneration of thoracolumbar intervertebral discs and late recurrence of clinical signs in dogs presented with acute thoracolumbar intervertebral disc extrusion and treated by hemilaminectomy alone. ANIMALS: Ninety-two client-owned dogs presented to 2 referral hospitals between 2009 and 2014. METHODS: Retrospective analysis of association between clinical signs consistent with recurrent thoracolumbar intervertebral disc extrusion and MRI evidence of disc degeneration in dogs undergoing hemilaminectomy for acute thoracolumbar intervertebral disc extrusion. Univariable and multivariable Cox regression analyses were used to explore associations between recurrence of clinical signs and several characteristics of T10-L3 discs at initial diagnosis. RESULTS: Ninety-two cases were included, of which 42 (46%) were Dachshunds and median age was 5.3 years. Clinical signs recurred in 33/92 (36%) dogs. Finding a completely degenerate disc in the T10 to L3 region (in addition to the operated site) at the time of surgery was associated with a hazard ratio of 2.92 (95% confidence interval: 1.37-6.20) for recurrence of clinical signs. CONCLUSIONS AND CLINICAL IMPORTANCE: Our results suggest that in cases of thoracolumbar intervertebral disc extrusion in dogs, recurrence of signs is likely if at least 1 completely degenerate disc in addition to the currently symptomatic disc is visible on MRI.

Comparison of survival after surgical or medical treatment in dogs with a congenital portosystemic shunt.

OBJECTIVE: To compare survival of dogs with a congenital portosystemic shunt (CPSS) that received medical or surgical treatment. DESIGN: Prospective cohort study. ANIMALS: 126 client-owned dogs with a single CPSS. PROCEDURES: Dogs were examined at 1 of 3 referral clinics, and a single CPSS was diagnosed in each. Dogs received medical or surgical treatment without regard to signalment, clinical signs, or results of hematologic or biochemical analysis. Survival data were analyzed via a Cox regression model. RESULTS: During a median follow-up period of 579 days, 18 of 126 dogs died as a result of CPSS. Dogs treated via surgical intervention survived significantly longer than did those treated medically. Hazard ratio for medical versus surgical treatment of CPSS (for the treatment-only model) was 2.9 (95% confidence interval, 1.1 to 7.2). Age at CPSS diagnosis did not affect survival. CONCLUSIONS AND CLINICAL RELEVANCE: Both medical and surgical treatment can be used to achieve long-term survival of dogs with CPSS, although results of statistical analysis supported the widely held belief that surgery is preferable to medical treatment. However, the study population consisted of dogs at referral clinics, which suggested that efficacy of medical treatment may have been underestimated. Although surgical intervention was associated with a better chance of long-term survival, medical management provided an acceptable first-line option. Age at examination did not affect survival, which implied that early surgical intervention was not essential. Dogs with CPSS that do not achieve acceptable resolution with medical treatment can subsequently be treated surgically.