JAVMA07-07-0367OSLevine0fig3tabVLSGJT

Adverse effects and outcome associated with dexamethasone administration in dogs with acute thoracolumbar disk herniation: 161 cases (20002006)
Jonathan M. Levine, dvm, dacvim; Gwendolyn J. Levine, dvm; Lindsay Boozer, bsa; Scott J. Schatzberg, dvm, phd, dacvim; Simon R. Platt, dvm, dacvim; Marc Kent, dvm, dacvim; Sharon C. Kerwin, dvm, dacvs; Geoffrey T. Fosgate, dvm, phd, dacvpm

SMALL ANIMALS/ AVIAN

ObjectiveTo determine complications and neurologic outcomes associated with dexamethasone administration to dogs with surgically treated thoracolumbar disk herniation, compared with dogs not receiving dexamethasone. DesignRetrospective case series. Animals161 dogs with surgically confirmed thoracolumbar disk herniation. ProceduresMedical records from 2 hospitals were used to identify dogs that had received dexamethasone < 48 hours prior to admission (dexamethasone group dogs), dogs that received glucocorticoids other than dexamethasone < 48 hours prior to admission (other-glucocorticoid group dogs), and dogs that received no glucocorticoids (nontreatment group dogs). Signalment, neurologic injury grade, laboratory data, and complications were extracted from medical records. ResultsDexamethasone group dogs were 3.4 times as likely to have a complication, compared with other-glucocorticoid or nontreatment group dogs. Dexamethasone group dogs were 11.4 times as likely to have a urinary tract infection and 3.5 times as likely to have diarrhea, compared with other-glucocorticoid or nontreatment group dogs. No differences in neurologic function at discharge or recheck evaluation were detected among groups. Conclusions and Clinical RelevanceResults indicate that treatment with dexamethasone before surgery is associated with more adverse effects, compared with treatment with glucocorticoids other than dexamethasone or no treatment with glucocorticoids, in dogs with thoracolumbar disk herniation. In this study population, no difference in outcome was found among groups. These findings suggest that the value of dexamethasone administration before surgery in dogs with thoracolumbar disk herniation should be reconsidered. (J Am Vet Med Assoc 2008;232:xxxxxx.)

ntervertebral disk herniation is the most common cause of acute spinal cord injury in dogs and according to the findings in 1 study1 leads to 2.3% of hospitalizations of all dogs.2,3 Primary and secondary spinal cord injury can occur as a result of disk herniation.46 Primary injury refers to the initial mechanical insult to the spinal cord and can consist of compression, concussion, contusion, and laceration.7 Secondary injury is the biochemical cascade that results from the primary insult and involves components of oxidative stress, excitotoxicity, inflammation, and vascular dysregulation.4,7 Surgical removal of herniated disk material often is performed in dogs with severe myelopathy, as it likely aids in relieving ongoing compression-related primary injury and may prevent the exacerbation of secondary processes. Treatment of secondary spinal cord injury is an
From the Departments of Small Animal Clinical Sciences (JM Levine, GJ Levine, Kerwin) and Veterinary Integrative Biosciences (Fosgate), College of Veterinary Medicine and Biomedical Sciences, Texas A&M University, College Station, TX 77843; and the Department of Small Animal Medicine and Surgery, College of Veterinary Medicine, University of Georgia, Athens, GA 30602 (Boozer, Schatzberg, Platt, Kent). Address correspondence to Dr. Jonathan Levine. JAVMA, Vol 232, No. 3, February 1, 2008

Abbreviations
TAMU UGA UTI CI Texas A&M University University of Georgia Urinary tract infection Confidence interval

active area of clinical and experimental research in veterinary and human medicine. High doses of glucocorticoids are the most widely used, commonly investigated, and frequently debated treatment for secondary injury.8 Experimental and clinical evidence exists supporting the use of high doses of glucocorticoids for acute spinal cord injury.9 High-dose glucocorticoid treatment is believed to contribute to neuronal protection principally by inhibiting lipid peroxidation, which may improve mitochondrial metabolism, preserve spinal cord blood flow, favorably alter ionic homeostasis, and reduce the amount of excitotoxic glutamate release.912 Results of various studies1317 on rats with spinal cord injury support the benefits of high-dose methylprednisolone and high-dose dexamethasone administration, showing improved motor recovery when treatment is provided soon
Scientific Reports 1

after trauma. Clinically derived evidence from doubleblinded, placebo-controlled studies1821 indicates that methylprednisolone sodium succinate,a if administered within 8 hours of injury, can improve motor outcome in humans with acute traumatic myelopathy. Despite numerous reports1821 extolling the benefits of high-dose glucocorticoid treatment in acute spinal cord injury, substantial data exist to suggest that these benefits may be limited, transient, or nonexistent in the clinical setting. Results of studies9,22,23 on experimentally induced spinal cord injury indicate that high doses of glucocorticoids can lead to excitotoxic neuronal death,22 worsening of oxidative injury through membrane phospholipase A2 inhibition,9 and lactate accumulation within the spinal cord parenchyma.23 Findings in several clinical studies2426 in humans with traumatic myelopathy have not revealed the purported benefits of high-dose methylprednisolone sodium succinate treatment. Results of a small retrospective study27 of dogs with intervertebral disk herniation did not reveal a positive benefit of high-dose methylprednisolone sodium succinate administration. Finally, numerous adverse effects are associated with high-dose glucocorticoid treatment in humans with spinal cord injury. including gastrointestinal hemorrhage, wound infection, and pneumonia.8 Data on dogs with spinal cord injury suggest that gastrointestinal signs, such as diarrhea and colonic perforation, are potential consequences of high-dose glucocorticoid treatment.2730 Although dexamethasone is used frequently in veterinary and human medicine as a treatment for acute spinal cord injury, information pertaining to its efficacy is limited, compared with treatment with other glucocorticoids such as methylprednisolone sodium succinate. The primary goal of the study reported here was to retrospectively evaluate the adverse effects associated with dexamethasone treatment in the immediate preoperative period in dogs with surgically addressed thoracolumbar disk herniation treated within 36 hours of hospitalization. A secondary goal was to obtain preliminary information concerning the efficacy of dexamethasone treatment in improving outcome in dogs with thoracolumbar disk herniation following surgery. It was our hypothesis that, compared with dogs that did not receive any glucocorticoids or received treatment with glucocorticoids other than dexamethasone, dogs that were provided dexamethasone would be more likely to have adverse effects associated with the gastrointestinal and urinary tracts. Materials and Methods Case selectionMedical records from dogs admitted to TAMU and the UGA veterinary medical hospitals between 2000 and 2006 with the diagnosis codes of lumbar disk herniation, thoracic disk herniation, and thoracolumbar disk herniation were reviewed. To be included in the study, dogs had to meet the following criteria. All dogs had to have a < 7 day history of neurologic signs at the time of initial evaluation at the universities. Findings on neurologic examination needed to suggest involvement of T3-L7 vertebrae and associated spinal cord segments. The diagnosis of disk herniation (protrusion or extrusion) had
2 Scientific Reports

SMALL ANIMALS/ AVIAN

to be confirmed by surgery, and all surgeries had to be performed within 36 hours of examination at the universities. It also was required that copies of referring veterinarian medical records generated 1 month prior to university hospital admission be available for review. Medical records reviewNo dogs in this study received glucocorticoids while hospitalized at either university. Dogs that were not provided any glucocorticoids within 1 month of admission to either university hospital were included in the nontreatment group. Dexamethasone group dogs were provided dexamethasone by referring veterinarians within 48 hours of university hospital admission and did not receive treatment with any other glucocorticoid within 1 month of university hospital admission. Dogs that received either methylprednisolone sodium succinate or prednisone at referring veterinary clinics within 48 hours of university hospital admission and that did not receive treatment with any other glucocoritcoids within 1 month of admission were included in the other-glucocorticoid group. Although no dosage requirements were followed for inclusion into the otherglucocorticoid and dexamethasone groups, the duration of glucocorticoid treatment was limited to a maximum of 48 hours on the basis of results of experimental and clinical studies21,23,31 indicating that administration of this duration can be effective in reducing the extent of spinal cord injury. ProceduresAge, sex, breed, weight, university (TAMU or UGA), duration of hospitalization, and dollar cost of hospitalization were obtained from the records for all dogs included in this study. A modified numerical Frankel spinal cord injury scale5,32,33 was used to retrospectively grade all dogs at initial evaluation and discharge from the universities as having paraplegia with no deep nociception (grade 0), paraplegia with no superficial nociception (grade 1), paraplegia with nociception (grade 2), nonambulatory paraparesis (grade 3), ambulatory paraparesis and ataxia (grade 4), spinal hyperesthesia only (grade 5), or no dysfunction (grade 6). Dexamethasone and other-glucocorticoid group dogs had the route of drug administration, dosage (mg/kg/d [mg/lb/d]), and number of days of treatment (1 or 2) recorded, although timing of drug delivery relative to date of client-reported onset of clinical signs was not examined because of variability in injury progression and the presence of ongoing injury throughout the interval before referral. Dexamethasone and other-glucocorticoid group dogs also had the type and duration of nonsteroidal anti-inflammatory treatment recorded, if this was provided. Complete blood count, serum biochemical analysis, blood lactate concentration, urinalysis, and urine bacterial culture data derived from initial evaluation at the universities were examined. The PCV serum albumin concen, tration, BUN concentration, and serum creatinine concentration were recorded. Hypernatremia (> 147 mEq of sodium/L) and hyperchloremia (> 116 mEq of chloride/ L), if present, were recorded. Blood lactate concentrationsb also were recorded when available (from TAMU only). Dogs were classified as having a UTI if they had > 5 WBCs/hpf and > 5 CFUs/mL of urine on urinalysis and urine bacterial culture results, respectively.
JAVMA, Vol 232, No. 3, February 1, 2008

Intensive care unit reports, discharge summaries, daily treatment logs, and daily subjective-objective assessments were examined for evidence suggesting vomiting, diarrhea, pneumonia, or wound infection during hospitalization. The diagnosis of wound infection was based on the admitting veterinarian assessment and did not require positive bacterial culture results. No attempt was made to categorize the number or severity of events. Statistical analysisData were summarized among the 3 treatment groups (dexamethasone, otherglucocorticoid, and nontreatment groups) by use of descriptive statistics. The Kruskal-Wallis test was used to compare medians of quantitative data among groups, and pairwise comparisons were performed by use of Mann-Whitney U tests with Bonferroni adjustment for multiple comparisons. Categoric variables were compared among the 3 treatment groups by use of 2 tests. Quantitative variables were compared between TAMU and UGA by use of Mann-Whitney U tests. Multivariable logistic regression was performed to identify factors significantly associated with possible adverse effects. Models were built independently for all adverse effects that were significant in the descriptive analysis (outcome) with the primary exposure of interest being whether the dog was treated with dexamethasone (vs all other dogs). Hospital (TAMU vs UGA) and modified Frankel score at admission were included in all models as potential confounders. The variables age (years), weight (kg [lb]), sex, neutering status (sexually intact vs neutered), and nonsteroidal anti-inflammatory administration (yes or no) were added to the model, and a backward stepwise procedure was used to remove variables 1 by 1 on the basis of conditional likelihood ratio tests. Within dexamethasone group dogs, the association between total dose and number of recorded adverse effects was assessed by use of the Pearson correlation coefficient. For comparisons and modeling, immediate improvement was defined as any dog that was discharged alive with a modified Frankel score greater than the value at presentation. Short-term improvement was defined as an improvement of 2 in the modified Frankel score be-

tween admission and the first recheck examination after surgery (typically 4 to 6 weeks after surgery). Proportions of immediate and short-term improvement were compared among groups by use of 2 tests. Multivariable logistic regression models were built to compare immediate and short-term improvement between dexamethasone group dogs, compared with other group dogs. Models were built in the same manner as described for the prediction of adverse effects. Continuous variables that did not satisfy the assumption of being linear in plots of the log-odds were included in logistic models as categoric variables. Hosmer and Lemeshow tests were calculated to assess the fit of models. The effect of dexamethasone treatment on the quantitative variables days in the hospital and client cost were modeled by use of a mixed linear approach with hospital included as a random effect and modified Frankel score at admission as a fixed effect to control for confounding. All analyses were performed by use of commercially available softwarec and interpreted at the 5% level of significance. Results A total of 1,003 records were searched at TAMU with 35 nontreatment, 43 dexamethasone, and 23 (17 received methylprednisolone sodium succinate and 6 received prednisone) other-glucocorticoid group dogs meeting inclusion criteria. A total of 1,005 records were searched at the UGA with 45 nontreatment, 6 dexamethasone, and 9 (6 received methylprednisolone sodium succinate and 3 received prednisolone) otherglucocorticoid group dogs meeting inclusion criteria. Mean SD age for dogs of all groups was 5.95 2.82 years (range, 1 to 15 years). A total of 161 dogs with 6 sexually intact females, 68 spayed females, 28 sexually intact males, and 59 male castrates were included. Breeds included Dachshund (miniature and standard grouped together; n = 87), mixed breed (14), Pekingese (11), Cocker Spaniel (8), Poodle (miniature and standard grouped together; 6), Shih Tzu (6), and 20 other breeds that included < 5 dogs each (29). No significant differences were found in age, sex, or breed among the treatment groups (Table 1). For dexamethasone group

SMALL ANIMALS/ AVIAN

Table 1Descriptive statistics and comparison of potential confounders for 161 dogs with confirmed disk herniation diagnosed at surgery in 2 veterinary referral institutions and treated with or without glucocorticoids by referring veterinarians during 2000 to 2006. Variable Age (y) Weight (kg) PCV (%) Albumin (mg/dL) BUN (mg/dL) No glucocorticoid Mean SD Min 5.9 2.7 11.3 9.7 51.7 6.0 3.74 0.45 14.1 5.8 2 3.3 31.0 2.6 5 Med 5 7.0 51.8a 3.6 13 Max 15 50.0 62.0 4.7 33 1.6 4.9 26 4 4,747 Other glucocorticoid* Mean SD Min 5.7 3.1 1 9.5 7.2 3.9 51.0 6.6 43.0 3.58 0.55 2.8 13.9 4.1 6 Med 5 7.4 49.5a,b 3.5 14 Max 13 36.4 67.0 5.1 21 1.2 11.2 19 4 4,695 Dexamethasone Mean SD Min Med Max 6.2 2.8 2 9.3 7.6 3.0 49.0 6.5 33.0 3.59 0.47 2.5 14.4 6.1 4 6 6.3 49.0b 3.6 13 14 36.8 64.0 4.8 33

Creatinine (mg/dL) 0.91 0.27 Lactate (mg/dL) 2.25 1.52 Hospital duration (d) 7.8 5.2 Frankel score at admission 2.2 1.2 Client cost ($) 2,388 607

0.5 0.8a 0.5 1.6 2 6a 0 2 1,101 2,378a

0.72 0.17 0.4 0.7b 4.36 3.34 1.0 3.7 6.6 4.8 1 6a,b 1.9 1.2 0 2 2,171 716 1,034 2,038b

0.78 0.41 0.5 0.7b 3.1 3.37 2.11 0.7 2.9 6.8 5.5 3.2 1 5b 17 2.1 1.2 0 2 4 2,287 548 1,434 2,203a,b 3,850

*Other glucocorticoids include prednisolone and methylprednisolone. P 0.05 for Kruskal-Wallis comparison of differences among medians. Min = Minimum. Med = Median. Max = Maximum. a,bMedians without superscripts in common are significantly (P 0.05) different based on Mann-Whitney U tests and Bonferroni adjustment for multiple comparisons. JAVMA, Vol 232, No. 3, February 1, 2008 Scientific Reports 3

dogs, mean SD total dose of dexamethasone was 2.25 4.28 mg/kg (1.02 1.95 mg/lb) with a range of 1 to 30 mg/kg (0.45 to 13.6 mg/lb). Modified Frankel scores at initial evaluation were not significantly (P = 0.56) different among groups, but the scores were significantly (P = 0.009) lower in dexamethasone group dogs at discharge, compared with admission. Treatment group dogs were significantly different in their admission date, with nontreatment group dogs being evaluated more recently (P < 0.001), compared with other group dogs. Client cost was significantly (P = 0.049) greater in nontreatment group dogs, although when adjusted for confounding by use of linear regression, this relationship was no longer significantly different among groups. Dexamethasone group dogs had shorter hospitalization duration, compared with nontreatment group dogs (7.8 days vs 5.5 days, respectively), but this finding was not significant when adjusted for confounding by hospital. Dogs treated at TAMU had significantly shorter hospital duration, compared with dogs treated at the UGA (Table 2). Serum biochemical analytes including serum albumin, BUN, serum sodium, and serum chloride concen-

trations were not different among groups. Serum creatinine concentration was higher in nontreatment group dogs (0.91 mg/dL), compared with other-glucocorticoid (0.72 mg/dL) and dexamethasone (0.78 mg/dL) group dogs. The mean blood lactate concentration in nontreatment group dogs was lower (2.25 mg/dL) than for dexamethasone (3.37 mg/dL) and other-glucocorticoid (4.36 mg/dL) group dogs, but this difference was not significant (P = 0.20). Median PCV in dexamethasone group dogs was lower than in nontreatment group dogs. A greater proportion of dexamethasone group dogs (3/49) were anemic, compared with other-glucocorticoid (0/32) or nontreatment group (2/78; PCV not recorded for 2 dogs) dogs, but this difference was not significant (Table 3). Dexamethasone group dogs (19/49) were more likely to have diarrhea than nontreatment group dogs (13/80). The proportion of dogs with vomiting or diarrhea was greater in the dexamethasone group (23/49) than the nontreatment group (17/80). Urinary tract infections also were more likely in dexamethasone group dogs (11/16; urine not evaluated for 33 dogs), com-

SMALL ANIMALS/ AVIAN

Table 2Descriptive statistics and comparison of potential confounders for 161 dogs with confirmed disk herniation diagnosed at surgery in 2 veterinary referral institutions and treated with or without glucocorticoids by referring veterinarians during 2000 to 2006. Variable Age (y) Weight (kg) PCV (%) Albumin (mg/dL) BUN (mg/dL) Creatinine (mg/dL) Lactate (mg/dL) Hospital duration (d) Frankel score at admission Client cost ($) Mean SD 6.1 3.0 8.6 6.6 50.3 6.6 3.45 0.33 13.7 5.5 0.76 0.33 3.13 2.31 5.6 3.3 2.2 1.2 2,351 577 1 3.0 33 2.5 4 0.4 0.5 1 0 1,101 TAMU Min Med 5 6.5 50 3.5 13 0.7 2.2 5 2 2,230 Max 15 36.8 67 4.1 33 3.1 11.2 17 4 4,695 Mean SD 5.7 2.6 13.3 10.7 51.4 5.9 4.11 0.44 15.3 5.7 0.96 0.25 NP 8.9 5.9 1.9 1.2 2,256 677 2 3.0 31 3.2 5 0.6 NP 2 0 1,034 UGA Min Med 5 8.3 51 4.2 15 0.9 NP 7 2 2,108 Max 13 50.0 66 5.1 27 1.6 NP 26 4 4,747 P Value* 0.406 0.002 0.353 0.001 0.051 0.001 NP 0.001 0.127 0.109

*P value based on Mann-Whitney U tests. NP = Not performed.

Table 3Comparison of immediate and short-term improvement and potential adverse effects for 161 dogs with confirmed disk herniation diagnosed at surgery in 2 veterinary referral institutions and treated with or without glucocorticoids by referring veterinarians during 2000 to 2006. Factor Immediate improvement Short-term improvement Vomiting Diarrhea Anemia High serum [Na] High serum [Cl] High serum [albumin] High blood [lactate] UTI Vomiting or diarrhea 1 adverse effect No glucocorticoid No. of dogs 78 37 80 80 78 65 64 53 13 15 80 80 No. of dogs with factor (%) 46 (59) 11 (30) 7 (9) 13 (16)a 2 (3) 25 (38) 8 (13) 15 (28) 4 (31) 2 (13)a 17 (21)a 53 (66)a Other glucocorticoid* No. of dogs 29 11 32 32 32 28 28 24 7 8 32 32 No. of dogs with factor (%) 15 (52) 3 (27) 7 (22) 6 (19)a,b 0 (0) 13 (46) 4 (14) 7 (29) 5 (71) 4 (40)a,b 12 (38)a,b 24 (75)a Dexamethasone No. of dogs 46 20 49 49 49 49 49 40 11 16 49 49 No. of dogs with factor (%) 19 (41) 7 (35) 6 (12) 19 (39)b 3 (6) 29 (59) 10 (20) 16 (40) 6 (55) 11 (69)b 23 (47)b 45 (92)b

*Other glucocorticoids include prednisolone and methylprednisolone. P 0.05 for 2 comparison of differences among percentages. [ ] = Concentration. a,bPercentages without superscripts in common are significantly (P 0.05) different based on pairwise 2 tests. 4 Scientific Reports JAVMA, Vol 232, No. 3, February 1, 2008

pared with the nontreatment group dogs (2/15; urine not evaluated for 65 dogs). Five of 161 dogs had wound infections after surgery (3 in the nontreatment group, 2 in the other-glucocorticoid group). No dogs had aspiration pneumonia. When all adverse effects were combined, 92% (45/49) of dexamethasone group dogs had 1 abnormalities, compared with 66% (53/80) of nontreatment group dogs. Higher doses of dexamethasone were not associated with more adverse effects (r = 0.183; P = 0.208). Multivariable logistic regression was used to estimate the association between dexamethasone treatment and possible adverse effects while adjusting for confounding by other variables included in the models. The modified Frankel score at admission did not satisfy the assumption of being linear in a plot of the log-odds and was therefore included in all logistic models as a categoric variable. Multivariable logistic models predicting the adverse effects of diarrhea, diarrhea and vomiting, UTI, and any 1 or more adverse effect fit the data well on the basis of the Hosmer and Lemeshow goodness-of-fit test with P values of 0.994, 0.689, 0.930, and 0.996, respectively. Dexamethasone group dogs were 3.5 (95% CI, 1.4 to 8.3) times as likely to develop diarrhea, compared with dogs in other groups, when adjusted for hospital, modified Frankel score at admission, sex, and neutering status. When vomiting and diarrhea were considered together, dexamethasone group dogs were 2.9 (95% CI, 1.3 to 6.4) times as likely to develop one of these adverse effects, compared with dogs in other groups, when adjusted for hospital, modified Frankel score at admission, and sex. Dexamethasone group dogs were 11.4 (95% CI, 1.5 to 88.8) times as likely to develop a UTI, compared with dogs in other groups. Female dogs were 18.5 (95% CI, 1.3 to 271) times as likely to develop a UTI, compared with male dogs. The multivariable model for the prediction of UTI adjusted the effects of hospital, modified Frankel score at admission, and nonsteroidal anti-inflammatory administration. Dexamethasone group dogs were 3.4 (95% CI, 1.0 to 11.2) times as likely to develop any adverse effect, compared with dogs in other groups. Dogs in all groups at TAMU were 5.9 (95% CI, 2.4 to 14.7) times as likely to have a report of adverse effects, compared with dogs at the UGA. Variables of hospital and dog weight also were included in the multivariable model predicting any 1 or more adverse effect. The proportion of dogs with immediate improvement (increase in modified Frankel score between admission and discharge; mean time, 6.9 days) in the dexamethasone group (19/46) was smaller than the proportion of dogs with immediate improvement in the nontreatment (46/78) or other-glucocorticoid (15/29) groups (Table 3), although this difference was not significant (P = 0.076). No significant differences were found in short-term outcome (increase in modified Frankel score by 2 between admission and reevaluation; mean follow-up, 29 days) with respect to groups. Discussion Since the 1960s, dexamethasone has been used as a treatment for acute spinal cord injury resulting from
JAVMA, Vol 232, No. 3, February 1, 2008

disk herniation in dogs, and it is currently recommended for this purpose by several investigators.3436 Dexamethasone can reduce the severity of traumatically induced edema and lipid peroxidation in the experimental setting, which has been a justification for its use in spinal cord injury.37,38 Early motor recovery has been enhanced in some spinal cord injury studies38,39 after administration of dexamethasone, although other investigators4042 have found no benefit associated with this intervention. Despite the frequent use of dexamethasone in dogs with injured spinal cords, few data exist concerning the frequency and type of adverse effects in this population or the efficacy of treatment in the clinical setting. High doses of glucocorticoids have been suggested to increase complications in animals and humans with endogenous spinal cord injury. The most frequently identified adverse effects in veterinary patients are gastrointestinal, although humans receiving methylprednisolone sodium succinate have increased rates of infection and aspiration pneumonia.4345 Overall rates of perioperative adverse effects seem to be higher in humans treated with high doses of glucocorticoids, compared with untreated control patients.43,44 In the study reported here, vomiting, diarrhea, UTI, wound infection, pneumonia, anemia, and various serum biochemical abnormalities associated with free water loss (high serum sodium, chloride, and albumin concentrations) were investigated with reference to glucocorticoid treatment. Dexamethasone group dogs were more likely to develop complications (45/49), compared with nontreatment group dogs (53/80), although the proportion of adverse effects (66%) was still substantial in the nontreatment group dogs. When multivariable logistic regression was used to adjust for confounding, dexamethasone group dogs were found to be 3.4 times as likely to develop adverse effects, compared with dogs in other groups. Complications associated with dexamethasone treatment were diarrhea and UTI in the study population of this report. Diarrhea was identified in 19 of 49 dexamethasone group dogs, compared with 13 of 80 nontreatment group dogs. Multivariable logistic regression analysis indicated that dexamethasone group dogs were 3.5 times as likely to develop diarrhea, compared with dogs in other groups. High-dose glucocorticoid administration can induce immunosuppression, gastric hyperacidity, and alterations in gastric mucosa, which, when coupled with preexisting spinal cord injury, may increase the risk of gastrointestinal complications.46 48 Dexamethasone group dogs also were 11.4 times as likely to have UTIs, compared with dogs in other groups. Although long-term glucocorticoid treatment has been recognized as a risk factor in the development of UTIs in pruritic dogs,49 short-term high-dose glucocorticoid administration has not, to our knowledge, been associated with UTI. Voiding disability resulting from spinal cord injury is well known to increase UTIs in dogs and may magnify the impact of altered bladder immune function that results from high-dose glucocorticoid administration.50,51 The inclusion of the Frankel score at admission in the multivariable model should have controlled for potential confounding by severity of disability for this comparison. Aspiration pneumoScientific Reports 5

SMALL ANIMALS/ AVIAN

nia and wound infection were not associated with dexamethasone administration. Few of the biochemical or hematologic variables assessed were different among groups after correcting for confounding, although further investigation may be needed. Dexamethasone group dogs did have a significantly lower PCV than the nontreatment group dogs, and a greater proportion of dexamethasone group dogs were anemic (3/49), compared with the other-glucocorticoid (0/32) or nontreatment group (2/78) dogs. Lactate concentrations were higher in the dexamethasone group dogs (3.37 mg/dL) than in the nontreatment group dogs (2.25 mg/dL), although these differences were not significant. The use of high doses of glucocorticoids in acute spinal cord injury is an ongoing controversy in human and veterinary medicine. Results of some studies18,21 indicate that humans with traumatic myelopathy may benefit from methylprednisolone sodium succinate, although the results of other studies8,25 have not supported the merit of this intervention. Results of 1 study27 in dogs with surgically confirmed thoracolumbar disk herniation did not reveal an improved outcome with methylprednisolone administration. In a study52 of 223 dogs with medically managed presumptive thoracolumbar disk herniation, glucocorticoid treatment was associated with reduced success and lower quality-of-life indices when logistic regression modeling was used to control for confounding variables. Any potential associations between outcome and treatment group in the study reported here must be interpreted with caution, as data were obtained retrospectively, substantial variability existed in drug delivery protocol (timing relative to spinal cord injury, duration of treatment, and dosage), and examiners likely differed in assessment techniques. Given the results of the study reported here and the limited information available from other experimental and clinical reports, we suggest that practitioners carefully consider the potential detriments and benefits of dexamethasone treatment prior to administering this drug to dogs with thoracolumbar disk herniation. Experimental evidence is controversial concerning effects on motor recovery after spinal cord injury, and clinical reports have yet to establish a substantial impact on outcome. Gastrointestinal signs and UTIs are potential complications associated with dexamethasone administration. Although death was not associated with adverse effects in the study reported here, clients should be warned about these risks prior to dexamethasone treatment. Additional retrospective and prospective studies are required to clarify the effects of dexamethasone in dogs with thoracolumbar disk herniation.
a. b. c. Solu-Medrol, Pharmacia, Kalamazoo, Mich. Critical Care Xpress, Nova Biomedical, Waltham, Mass. SPSS, version 14.0, SPSS Inc, Chicago, Ill.

4. 5.

6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

16. 17.

18.

19.

20. 21.

22. 23. 24.

References
1. 2. 3. Hoerlein BF Canine neurology: diagnosis and treatment. Philadel. phia: WB Saunders, 1978. Dewey CW. Myelopathies: disorders of the spinal cord. In: Dewey CW, ed. A practical guide to canine and feline neurology. Ames, Iowa: Iowa State Press, 2003;277336. Coates JR. Intervertebral disk disease. Vet Clin North Am Small Anim Pract 2000;30:77110. Scientific Reports

25. 26. 27.

Olby N. Current concepts in the management of acute spinal cord injury. J Vet Intern Med 1999;13:399407. Levine JM, Ruaux CG, Bergman RL, et al. Matrix metalloproteinase-9 activity in the cerebrospinal fluid and serum of dogs with acute spinal cord trauma from intervertebral disk disease. Am J Vet Res 2006;67:283287. McMichael MA, Ruaux CG, Baltzer WI, et al. Concentrations of 15F2t isoprostane in urine of dogs with intervertebral disk disease. Am J Vet Res 2006;67:12261231. McDonald JW, Sadowsky C. Spinal-cord injury. Lancet 2002; 359:417425. Hurlbert RJ. Methylprednisolone for acute spinal cord injury: an inappropriate standard of care. J Neurosurg 2000;93:17. Hall ED, Springer JE. Neuroprotection and acute spinal cord injury: a reappraisal. NeuroRx 2004;1:80100. Hall ED. Lipid antioxidants in acute central nervous system injury. Ann Emerg Med 1993;22:10221027. Hall ED. The neuroprotective pharmacology of methylprednisolone. J Neurosurg 1992;76:1322. Taoka Y, Okajima K, Uchiba M, et al. Methylprednisolone reduces spinal cord injury in rats without affecting tumor necrosis factor-alpha production. J Neurotrauma 2001;18:533543. Sharma A, Tiwari R, Badhe P et al. Comparison of methylpred, nisolone with dexamethasone in treatment of acute spinal injury in rats. Indian J Exp Biol 2004;42:476480. Delattre JY, Arbit E, Rosenblum MK, et al. High dose versus low dose dexamethasone in experimental epidural spinal cord compression. Neurosurgery 1988;22:10051007. Sasaki T, Sakuma J, Ichikawa T, et al. Effects of methylprednisolone on axonal depression induced by hypoxia, gammaaminobutyric acid, and (+/)-8-hydroxy-dipropylaminotetralin hydrobromide. Neurosurgery 2002;51:14771483. Brandoli C, Shi B, Pflug B, et al. Dexamethasone reduces the expression of p75 neurotrophin receptor and apoptosis in contused spinal cord. Brain Res Mol Brain Res 2001;87:6170. Cherian L, Kuruvilla A, Chandy MJ, et al. Changes in phospholipids and acetylcholinesterase during early phase of injury to spinal cordan experimental study in rats. Indian J Physiol Pharmacol 1996;40:134138. Bracken MB, Shepard MJ, Collins WF et al. A randomized, con, trolled trial of methylprednisolone or naloxone in the treatment of acute spinal-cord injury. Results of the Second National Acute Spinal Cord Injury Study. N Engl J Med 1990;322:14051411. Bracken MB, Shepard MJ, Collins WF Jr, et al. Methylprednisolone or naloxone treatment after acute spinal cord injury: 1year follow-up data. Results of the second National Acute Spinal Cord Injury Study. J Neurosurg 1992;76:2331. Bracken MB. Pharmacological treatment of acute spinal cord injury: current status and future projects. J Emerg Med 1993; 11(suppl 1):4348. Bracken MB, Shepard MJ, Holford TR, et al. Administration of methylprednisolone for 24 or 48 hours or tirilazad mesylate for 48 hours in the treatment of acute spinal cord injury. Results of the Third National Acute Spinal Cord Injury Randomized Controlled Trial. National Acute Spinal Cord Injury Study. JAMA 1997;277:15971604. Saplosky RM. Glucocorticoids, stress and exacerbation of excitotoxic neuron death. Semin Neurosci 1994;6:323331. Braughler JM, Hall ED. Lactate and pyruvate metabolism in injured cat spinal cord before and after a single large intravenous dose of methylprednisolone. J Neurosurg 1983;59:256261. McCutcheon EP Selassie AW, Gu JK, et al. Acute traumatic spi, nal cord injury, 19932000. A population-based assessment of methylprednisolone administration and hospitalization. J Trauma 2004;56:10761083. Pollard ME, Apple DF Factors associated with improved neu. rologic outcomes in patients with incomplete teraplegia. Spine 2003;28:3338. Pointillart V, Petitjean ME, Wiart L, et al. Pharmacological therapy of spinal cord injury during the acute phase. Spinal Cord 2000;38:7176. Boag AK, Drobatz KJ. Complications of methylprednisolone sodium succinate in dachshunds with surgically treated intervertebral disc disease. J Vet Emerg Crit Care 2001;11:105110. JAVMA, Vol 232, No. 3, February 1, 2008

SMALL ANIMALS/ AVIAN

28. Toombs JP, Caywood DD, Lipowitz AJ, et al. Colonic perforation following neurosurgical procedures and corticosteroid therapy in four dogs. J Am Vet Med Assoc 1980;177:6872. 29. Moore RW, Withrow SJ. Gastrointestinal hemorrhage and pancreatitis associated with intervertebral disk diseases in the dog. J Am Vet Med Assoc 1982;180:14431447. 30. Culbert LA, Marino DJ, Baule RM, et al. Complications associated with high-dose prednisolone sodium succinate therapy in dogs with neurological injury. J Am Anim Hosp Assoc 1998;34:129134. 31. Anderson DK, Saunders RD, Demediuk P, et al. Lipid hydrolysis and peroxidation in injured spinal cord: partial protection with methylprednisolone or vitamin E and selenium. Cent Nerv Syst Trauma 1985;2:257267. 32. Levine JM, Levine GJ, Kerwin SC, et al. Association between various physical factors and acute thoracolumbar intervertebral disk extrusion or protrusion in Dachshunds. J Am Vet Med Assoc 2006;229:370375. 33. Frankel HL, Hancock DO, Hyslop G, et al. The value of postural reduction in the initial management of closed injuries of the spine with paraplegia and tetraplegia. Paraplegia 1969;7: 179192. 34. Kraus KH. Medical management of acute spinal cord disease. In: Bonagura JD, ed. Kirks current veterinary therapy XIII. Philadelphia: WB Saunders Co, 1999;186190. 35. Plumb DC. Veterinary drug handbook. 4th ed. Ames, Iowa: Iowa State Press, 2002. 36. Wilcox KR. Conservative treatment of thoracolumbar intervertebral disc disease in the dog. J Am Vet Med Assoc 1965;147: 14581460. 37. Braughler JM. Lipid peroxidation-induced inhibition of gammaaminobutyric acid uptake in rat brain synaptosomes: protection by glucocorticoids. J Neurochem 1985;44:12821288. 38. Kuchner EF Hansebout RR, Pappius HM. Effects of dexametha, sone and of local hypothermia on early and late tissue electrolyte changes in experimental spinal cord injury. J Spinal Disord 2000;13:391398. 39. De La Torre JC, Johnson CM, Goode DJ, et al. Pharmacologic treatment and evaluation of permanent experimental spinal cord trauma. Neurology 1975;25:508514. 40. Hoerlein BF Redding RW, Hoff EJ, et al. Evaluation of dexa,

41. 42. 43. 44.

45.

46. 47.

48. 49. 50.

51. 52.

methasone, DMSO, mannitol and solcosyl in acute spinal cord trauma. J Am Anim Hosp Assoc 1983;19:216225. Faden AI, Jacobs TP Patrick DH, et al. Megadose corticosteroid , therapy following experimental traumatic spinal injury. J Neurosurg 1984;60:712717. Rucker NC, Lumb WV, Scott RJ. Combined pharmacologic and surgical treatments for acute spinal cord trauma. Am J Vet Res 1981;42:11381142. Gerndt SJ, Rodriguez JL, Pawlik JW, et al. Consequences of high-dose steroid therapy for acute spinal cord injury. J Trauma 1997;42:279284. Molano MD, Broton JG, Bean JA, et al. Complications associated with the prophylactic use of methylprednisolone during surgical stabilization after spinal cord injury. J Neurosurg 2002;96(suppl 3):S267S272. Matsumoto T, Tamaki T, Kawakami M, et al. Early complications of high-dose methylprednisolone sodium succinate treatment in the follow-up of acute cervical spinal cord injury. Spine 2001;26:426430. Britt RC, Devine A, Swallen KC, et al. Corticosteroid use in the intensive care unit. At what cost? Arch Surg 2006;141:145149. Rohrer CR, Hill RC, Fischer A, et al. Efficacy of misoprostol in prevention of gastric hemorrhage in dogs treated with high doses of methylprednisolone sodium succinate. Am J Vet Res 1999;60:982985. Rohrer CR, Hill RC, Fischer A, et al. Gastric hemorrhage in dogs given high doses of methylprednisolone sodium succinate. Am J Vet Res 1999;60:977981. Torres SM, Diaz SF Nogueira SA, et al. Frequency of urinary tract , infection among dogs with pruritic disorders receiving long-term glucocorticoid treatment. J Am Vet Med Assoc 2005;227:239243. Stiffler KS, Stevenson MA, Sanchez S, et al. Prevalence and characterization of urinary tract infections in dogs with surgically treated type 1 thoracolumbar intervertebral disc extrusion. Vet Surg 2006;35:330336. Seguin MA, Vaden SL, Altier C, et al. Persistent urinary tract infections and reinfections in 100 dogs (19891999). J Vet Intern Med 2003;17:622631. Levine JM, Levine GJ, Johnson SI, et al. Evaluation of the success of medical management for presumptive thoracolumbar intervertebral disk herniation in dogs. Vet Surg 2007;36:482491.

SMALL ANIMALS/ AVIAN

JAVMA, Vol 232, No. 3, February 1, 2008

Scientific Reports