Retrospective evaluation of hyperproteinorrachia without pleocytosis (albuminocytologic dissociation) and survival in dogs

This is an open access article under the terms of the Creat ive Commo ns Attri bution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2021 The Authors. Veterinary Clinical Pathology published by Wiley Periodicals LLC on behalf of American Society for Veterinary Clinical Pathology 1Queen’s Veterinary Hospital, University of Cambridge, Cambridge, UK 2Millennium Veterinary Practice, Braintree, UK


| INTRODUC TI ON
Hyperproteinorrachia without pleocytosis (HP) (also known as albuminocytologic dissociation) is defined by raised microprotein concentrations within cerebrospinal fluid (CSF) 1 and can be seen in dog CSF samples analyzed by veterinary clinical pathologists. The increase in protein is attributed to either (a) local intrathecal immunoglobulin production (eg, seen with inflammatory or neoplastic diseases) or neurologic tissue-derived proteins (eg, neurodegeneration leading to the release of cytoskeletal proteins), (b) increased bloodcentral nervous system (CNS) barrier permeability allowing bloodderived proteins to enter the CSF, or (c) interruption of CSF flow or absorption or, as is most likely, a combination of these scenarios. 2 The term albuminocytologic dissociation has not been used here as it does not accurately capture the associated pathophysiology leading to increased CSF total protein concentrations (CSF-TP); instead, it infers that hyperproteinorrachia is solely defined by albumin concentration increases. Many different disease processes can result in these circumstances; therefore, HP has a low specificity 1,2 and, recent consolidated literature regarding HP in veterinary species is lacking. As a consequence, receiving a diagnosis of HP is largely descriptive and frequently not informative for veterinarians.
The term albuminocytologic dissociation (HP without pleocytosis) was first coined in 1912, describing the phenomenon in human patients with spinal cord compression. 3 In human medicine, this finding is now associated with a wide variety of inflammatory (eg, Guillain-Barrè syndrome [GBS]), infectious (eg, viral meningitis), degenerative, and neoplastic conditions. 4 Recently, the importance of age-adjusted CSF-TP reference intervals has been described (with a small sex effect also reported), as human CSF-TP concentrations naturally elevate with increasing age, thus reducing the number of false-positive HP diagnoses. 4 The same age-(and sex-) association has not been widely reported within the veterinary literature. The Historically, canine CSF protein electrophoresis was conducted to quantify albumin and immunoglobulin proportions helping to elucidate underlying disease mechanisms at play, thus increasing specificity. 6 This practice is now largely redundant with the advent of advanced diagnostics (eg, advanced imaging, infectious disease testing), allowing the diagnosis of specific neurologic conditions. However, not all cases can be extensively tested (eg, due to economic constraints), and advanced imaging has specificity limitations (specifically distinguishing inflammatory from neoplastic lesions); therefore, if more recently published, consolidated data regarding HP-associated conditions and factors associated with survival were available, clinical applications would be more easily identifiable.
Consequently, we aimed to (a) identify the most common conditions associated with HP within a UK population of dogs, (b) investigate whether a higher CSF-TP concentration is associated with reduced survival times, and (c) identify if any other risk factors in patients with HP are associated with survival times.

| Data acquisition
Ethical approval for this study was obtained from the Department Patient data (case number, species, breed, and sex) from the HP cases were collated, and presenting clinical examination, neuroanatomic localization (at admission), clinical pathology data, and definitive diagnosis (if reached) were recorded. A definitive diagnosis was made by the attending board-certified neurologist using all available patient data, including advanced imaging investigation (magnetic resonance imaging) and/or confirmation with histopathology. After collection, all data were anonymized.
Patient survival (defined as being alive 1 year after the analysis date for the CSF sample that showed HP) was recorded for each individual, and initially, information was obtained from patient records.
Where this information was not present, the referring veterinary surgeon was contacted.

| Statistical analysis
Statistical analyses were performed using either GraphPad Prism age, CSF-TP and corrected CSF-TP, weight, temperature, heart and respiratory rate at presentation, and categorical variables, breed (stratified as breed types, such as mixed breed and terrier-type), sex, neurologic localization, the presence of inflammatory changes on hematology and/or biochemistry panels (defined as an inflammatory leukogram or increased positive acute-phase proteins, respectively), and disease classification. If the assumptions for the Cox proportional hazard model were violated, the n number was too low, or all individuals within that group were censored then the group was excluded from analysis. Schoenfeld residuals were calculated for each variable to ensure none violated the assumptions of proportional hazards.

| Diseases associated with HP in dogs
From the records (748 CSF results with a microprotein value), 39 dogs with HP were identified. An additional 14 dogs with HP were excluded due to hemodilution of the sample. The presence of HP was associated with 17 different conditions ( Table 1), and of those with neoplasia (n = 6), the exact neoplasm was undetermined in three patients. Intervertebral disc disease and meningoencephalitis of unknown origin (MUO) were the most commonly associated conditions. Definitive diagnosis of neoplasia was obtained in 4/6 patients through histologic examinations (either ante-mortem biopsies or post-mortem examination), including hemangiosarcoma (n = 1) and peripheral nerve sheath tumor (n = 3). Neoplasia was suspected in two others (lymphoma and one undetermined).

| CSF-TP in patients with HP
Descriptive data are summarized in Table 2A

| CSF-TP and survival in patients with HP
Complete patient survival at 1 year post-HP identification was available for 21 patients, and within this group there was no difference in the CSF-TP or corrected CSF-TP concentrations between  The univariable analysis identified neoplasia as being negatively correlated with survival (P < 0.05, Hazard Ratio (HR): 11.07) (Table 2B) when the non-neoplastic diseases were combined and compared with neoplastic diseases.
Increasing age is associated with the development of some types of neoplasia within the nervous system. 8    No statistically significant differences were noted between any of the group comparisons. Dots reflect individual patient values, with the mean and standard deviation. DAMNITV, degenerative, anomalous, metabolic, neoplastic, iatrogenic, idiopathic, inflammatory/immune-mediated, toxic, traumatic, or vascular This is also the first paper evaluating the survival to 1 year of dogs with HP irrespective of the underlying etiopathogenesis or condition; survival in MUO patients was not reported to be associated with CSF-TP concentrations. 9 No correlation between CSF-TP concentrations (including when correct for the sampling site) and survival was identified, and while this might be a true reflection of HP, the heterogeneity of the conditions included could also have influenced this lack of a correlation. Notwithstanding, grouping the conditions into general categories of disease etiology (eg, degenerative conditions) did not highlight any significant differences in CSF-TP that could have been influenced by the inclusion of other conditions. Indeed, neoplastic conditions were associated with the poorest 1-year survival, and these HP cases typically had only modest CSF-TP elevations (Figures 2 and 3). However, expanding this study to include more individuals from the most common diseases present would have allowed more disease-specific statements to be made. The three main limitations of this research include the retrospective nature of the study, the relatively low number of dogs included, and, potentially, differences in sample acquisition times between individuals and the stage of disease. The archives were searched over an 11-year period, and only approximately 5% of the CSF samples with a CSF-TP result were identified as having HP. This low number also precluded disease-specific survival analysis. Therefore, multicenter studies would be needed to expand case numbers, which could allow further analyses to be performed.

| D ISCUSS I ON
The timing of sample acquisition was not standardized across the HP cases included, as this is individually recommended based on disease presentation. Consequently, it means that some dogs may have been sampled during the acute phase of the underlying disease, whereas others could have been sampled during the chronic or resolving phases. As the aim of this study was to investigate different conditions that result in HP and the relationship of HP to survival, the lack of a standardized CSF sampling time point is likely less important than if a sole condition associated with HP was being investigated.
In conclusion, HP in dogs is associated with a wide range of conditions, most commonly neoplasia, MUO, and IVDD. The higher CSF-TP concentrations did not correlate with a worse 1-year survival; however, neoplastic lesions do. Increasing age in dogs does not appear to be associated with increasing CSF-TP concentrations.

D I SCLOS U R E
The authors have indicated that they have no affiliations or financial involvement with any organization or entity with a financial interest in, or in financial competition with, the subject matter or materials discussed in this article.