Criteria for Selection of Cases

Medical records of dogs with tetanus were identified retrospectively by searching for problems or clinical diagnoses classified as tetanus or tetany in the computer database of the Veterinary Medical Teaching Hospital at the University of California, Davis, for the years 1987 through 2005. A dog was excluded if it had confirmed strychnine or other neurotoxicosis; serum ionized, total, or (when applicable) corrected calcium concentration less than the lower reference limit; confirmed extraocular, masticatory, or other myositis; rigidity of a single limb as the only abnormality; or if the medical record was incomplete or missing. A dog was included in the study if it had typical signs of local or generalized tetanus, including retraction of lip commissures (risus sardonicus), erect ears, trismus, or rigidity of multiple limbs.

Procedures

The medical records of identified dogs were reviewed, and information was collected by use of a standardized data sheet from those that were eligible for inclusion in the study. The information recorded included signalment; clinical signs noted by the owner on the first day of illness; day of first evaluation by a veterinarian (primary or referral); initial complaint and physical examination findings; wound characteristics and microbiologic culture results; physical examination findings on the day of worst clinical signs; abnormal electrocardiographic and blood pressure measurements; details regarding treatment with tetanus antitoxin, antimicrobial administration, wound management technique, nutritional plan, and use of sedatives; complications; and outcome. The first day of illness was designated day 0; survival or nonsurvival at day 28 was recorded.

For the purposes of this study, hyperthermia was defined as rectal temperature ≥ 39.4°C (≥ 103.0°F). Bradycardia was defined as heart rate ≤ 60 beats/min, and tachycardia was defined as heart rate ≥ 140 beats/min. Hypotension was defined as mean arterial pressure ≤ 60 mm Hg or systolic blood pressure d 80 mm Hg, and hypertension was defined as mean arterial pressure ≥ 130 mm Hg or systolic blood pressure ≥ 150 mm Hg. Bradypnea was defined as respiratory rate ≤ 10 breaths/min and tachypnea was defined as respiratory rate ≥ 40 breaths/min without panting.

Severity classification—To simplify reporting of the clinical course of the disease, a canine tetanus severity classification system similar to the tetanus severity classification system most commonly used in humans^28–30 was developed (Appendix). The system was developed prior to identification of dogs eligible for inclusion in the study and was based on the human classification system, information gathered from veterinary textbooks and case reports of dogs with tetanus, and the authors' clinical experience. In the resultant tetanus severity classification system, class I included dogs with only facial signs of tetanus; class II included dogs with generalized rigidity or dysphagia, with or without class I signs; class III included dogs with class I or II signs that were recumbent or had seizures; and class IV include dogs with class I, II, or III signs as well as abnormal heart rate, respiratory rate, or blood pressure measurements.

Statistical analysis—Relationships between categoric variables were analyzed by use of the C² test of homogeneity. Differences in continuous measurements across ordered population categories were determined by use of the Jonckheere-Terpstra test. Relationships between continuous or ordinal variables were analyzed by use of the Spearman rank correlation. The effect of variables on the rate of death was determined via Cox proportional hazards regression and reported as hazard ratios with 95% CIs. Dogs that died for reasons unrelated to tetanus and those lost to follow-up were censored from the survival analyses. A value of P < 0.05 was considered significant.

Results

Fifty-nine records of dogs with tetanus or tetany were identified; 21 dogs were excluded because of hypocalcemia (n = 11), strychnine toxicosis (3), incomplete medical record (2), clinical signs that did not match the inclusion criteria (2), a single rigid limb (1), metaldehyde toxicosis (1), and extraocular myositis (1). Thirtyeight dogs were included in the study. Breeds included Labrador Retriever or Labrador mix (n = 5), Husky or Husky mix (4), Cocker Spaniel (3), Dachshund (3), Australian Cattle Dog (3), Golden Retriever or Golden Retriever mix (3), English Springer Spaniel (2), Rottweiler (2), Doberman Pinscher (2), Airedale Terrier (1), Boxer (1), Catahoula Hog Dog (1), German Shepherd Dog (1), German Shorthaired Pointer (1), Great Dane (1), Old English Sheepdog (1), Pit Bull Terrier (1), Samoyed (1), Weimaraner (1), and mixed breed (1). Among the dogs, 4 were sexually intact females, 12 were spayed females, 7 were sexually intact males, and 15 were castrated males. The dogs' ages ranged from 8 weeks to 14 years (median age, 4 years). Younger dogs were significantly (P = 0.04) more likely to develop a worse tetanus severity class than older dogs. Although it appeared that the rate of death decreased by 0.78% (hazard ratio, 0.78; 95% CI, 0.58 to 1.06) for every year of age, this relationship was nonsignificant (P = 0.11).

Clinical course of disease—Clinical signs detected on day 0 by the owners were recorded for all dogs. The most common clinical signs on day 0 were ocular changes, specifically third eyelid protrusion or sunken eyes (n = 11), an odd appearance of the eyes (as described by the owner; 4), or squinting or swollen eyelids (3); one of the dogs had both squinted eyelids and a sunken appearance of the eyes. Frequently, ocular signs were the only initial abnormality detected by the owner. Eight dogs had tense muscles of the face or neck; 3 other dogs had facial swelling. On day 0, clinical signs in 6 dogs included difficulty eating or dropping food; 1 other dog had hypersalivation without definite dysphagia. For 7 dogs, the owners described episodes of apparent regurgitation or vomiting, although it was generally unclear which had occurred. Only 4 dogs had limb stiffness initially, and 2 other dogs were found laterally recumbent, unable to stand, and trembling. One dog had a voice change and 1 was dribbling urine. In 6 dogs, the only abnormality noted on day 0 was lethargy or anorexia.

The day of first evaluation by a veterinarian was recorded for 37 dogs. Dogs were examined by a veterinarian on days 0 through 5 (median, day 2). In most instances, clinical signs had progressed from those detected by the owner on day 0 by the time the dog was evaluated by a veterinarian. At the first evaluation, 16 dogs fit the criteria for tetanus severity class I, 17 dogs fit the criteria for class II, and 3 dogs fit the criteria for class III; no dogs fit the criteria for class IV at the first evaluation. For 1 dog, the owner's only complaint was anorexia and there were no abnormal findings on physical examination; therefore, it was not initially classified with the canine tetanus severity classification system. Nine dogs that were initially graded as class I and for which the initial complaint did not definitively include dysphagia were regurgitating or vomiting food on the day of the first evaluation.

Most dogs developed more severe clinical signs after the first evaluation. Of the 16 dogs initially graded as class I severity, 4 dogs remained in this classification, 6 progressed to class II, 4 progressed to class III, and 2 progressed to class IV. Of the 17 dogs initially graded as class II severity, 9 dogs remained in this classification, 1 progressed to class III, and 7 progressed to class IV. The 3 dogs initially graded as class III severity all progressed to class IV. Both the dog for which initial physical examination findings were not recorded and the dog with apparently normal initial physical examination findings progressed to class IV. Therefore, the highest recorded tetanus severity class assigned was class I for 4 dogs, class II for 15 dogs, class III for 5 dogs, and class IV for 14 dogs. The number of dogs in each tetanus severity class with specific clinical signs was assessed (Table 1).

Four dogs graded as tetanus severity class IV were unable to urinate voluntarily and required catheterization. One dog each in classes II, III, and IV was constipated and received enemas. A hiatal hernia was diagnosed radiographically in 1 dog in class III and 2 dogs in class IV. One of these dogs was euthanized because of severe tetany while the hernia was still present, 1 underwent surgical hernia reduction during placement of a gastrostomy tube for nutritional support, and 1 had spontaneous hernia resolution after recovery from tetanus. Tetanic spasms resulted in hyperthermia (rectal temperatures, 39.6° to 43.9°C [103.2° to 111.0°F]) in 4 dogs in class III and 11 dogs in class IV. Spastic muscle rigidity caused such severe hypoventilation in 2 dogs in class IV that anesthesia and mechanical ventilation were required; a third dog died as a result of apnea during one of these episodes before mechanical ventilation could be instituted. Six dogs had moderate to severe dyspnea during tetanic spasms.

Twelve dogs had moderate bradycardia (heart rate, 40 to 60 beats/min), and in some, bradycardia was episodic. Two of the dogs with bradycardia had second-degree atrioventricular block, whereas a third had bradyarrhythmia with dropped heart beats detected via cardiac auscultation (no ECG was performed). In 1 dog, the bradycardic episodes required the placement of a temporary pacemaker, from which the dog was weaned as the tetanus resolved. Eight dogs had episodes of tachycardia (heart rate, 150 to 210 beats/min); 6 of these dogs also had intermittent bradycardia. One of these 8 dogs also had ventricular premature complexes, and another also had ventricular premature complexes and periods of accelerated idioventricular rhythm. Blood pressure was measured in 6 dogs. The 3 dogs with abnormal blood pressure measurements also had bradycardia, 2 had episodic hypertension, and 1 had episodes of severe hypotension. It was impossible to discern from some medical records whether episodes of tachycardia and hypertension occurred only during tetanic spasms, or whether those abnormalities occurred spontaneously. Thus, no so-called autonomic storms were positively identified. One dog with intermittent tachycardia and the dog with episodic hypotension had SIRS. One dog with intermittent bradycardia and tachycardia also had periods of apnea that occurred independent of tetanic spasms; the cause of these apneic events was unknown.

Antitoxin—Information regarding antitoxin administration was available for 37 dogs. Twenty-nine dogs were treated with tetanus antitoxin. The day of antitoxin administration was recorded for 28 dogs; median day of administration was day 3 (range, day 0 to day 13). There was no association between earlier administration of antitoxin and progression of clinical signs (ie, worsening tetanus severity classification) or 28-day mortality rate. The dose of antitoxin administered was known for 28 of the treated 29 dogs; doses ranged from 10 to 1,900 U/kg (4.5 to 863.6 U/lb), and the median value was 326.5 U/kg (148.4 U/lb). Results of skin tests before antitoxin administration were available for 19 dogs; test results were negative for 17 dogs and positive for 2 dogs. Both dogs with positive skin test results were premedicated before antitoxin administration (one with diphenhydramine and the other with diphenhydramine and dexamethasone); neither dog had a reaction to antitoxin administration. During IV infusion of antitoxin, reactions to antitoxin were recorded for 4 of the 29 treated dogs. In all 4 dogs, vomiting or retching occurred, and 1 of these dogs had tachypnea. Among the 4 dogs that developed an adverse reaction to antitoxin, the infusion rate was slowed for 2, discontinued for 1, and discontinued for 1 in addition to administration of a dose of dexamethasone. All dogs recovered from the adverse reactions as a result of these measures. Three of the dogs that reacted to antitoxin had yielded negative skin test results and had been premedicated with diphenhydramine, and 1 dog had not undergone skin testing and had not been premedicated.

Antimicrobials—Information regarding antimicrobial administration was available for 35 dogs. Penicillin G or metronidazole was considered an appropriate antimicrobial for the treatment of C tetani infection. Thirty-three dogs were treated with appropriate antimicrobials. Twenty-one dogs were treated with metronidazole, 2 dogs were treated with penicillin, and 10 dogs were treated with both agents. The day of initiation of antimicrobial treatment was recorded for 30 of the 35 dogs, which ranged from days 0 to 10 (median, day 3). There was no significant association between earlier antimicrobial administration and progression of clinical signs (ie, worsening tetanus severity classification) or 28-day mortality rate.

Wounds and wound management—Records indicated that a wound was evident in 28 dogs; the remaining 10 dogs had no known injury and no visible wound on examination. The day of wound occurrence was known in 11 dogs and ranged from days −18 to −3 (median, day −9). In 17 dogs for which the day of wounding was not definitively known, a wound was detected from 8 days before to 5 days after onset of clinical signs (median, day −4). Twenty-two dogs had external wounds, including draining wounds, abscesses, or healing wounds of the feet or claws (n = 10); draining tracts associated with grass awns (6); facial wounds or swelling (3); draining tract on the lateral aspect of the thorax (1); multiple tick bite wounds after being lost in the woods for 6 weeks (1); and bleeding claw wounds from excessive trimming (1). Six dogs had surgical wounds as result of ovariohysterectomy (n = 3), castration (1), perineal hernia repair (1), and skin flap surgery after extensive dog bite wounds (1). Of the 10 dogs in which no wound was identified, 2 had frequent exposure to grass awns and 2 others were teething. Dogs with an internal or surgical wound appeared to be more likely to progress to a higher tetanus severity class than dogs with external wounds, but this association was not significant (P = 0.091). There was no association between wound type and 28-day mortality rate.

Among the 28 dogs with detectable wounds, 14 received wound treatment. External wounds were treated in 11 of 22 dogs. In 2 of the 3 dogs with tetanus that had undergone ovariohysterectomy, exploratory laparotomy with stump debridement was performed. The dog that had undergone skin flap surgery received surgical wound management. The other 3 dogs with surgical wounds did not undergo surgical wound debridement after the diagnosis of tetanus was made. Wounds were treated from days 0 to 17 (median, day 4). There was no significant association between earlier wound management and progression of clinical signs (ie, worsening tetanus severity classification) or 28-day mortality rate. Anaerobic bacterial cultures of wound material were completed for 5 dogs, and C tetani was not isolated from any of those samples.

Nutrition—Nutritional information was available for 29 of 31 dogs that were treated as hospital inpatients. Of those 29 dogs, 10 received a PEG tube. Three dogs received TPN alone, 2 dogs were treated with a combination of TPN and a PEG tube, and 1 dog was treated with TPN and a surgically placed gastrostomy tube. Ten dogs were fed orally with either gruel or meatballs. One dog was managed with an esophagostomy tube, and 2 dogs did not receive nutritional support during hospitalization. Two dogs had complications associated with nutritional support. The dog with the surgically placed gastrostomy tube developed vomiting and required nutritional support with TPN until the tube could be used; another dog dislodged its PEG tube while its clinical signs of tetanus were improving, resulting in septic peritonitis and subsequent euthanasia.

Muscle relaxation—Twenty-six dogs were treated with sedatives during hospitalization. All dogs treated with sedatives were in tetanus severity classes II, III, and IV, and multiple drugs were administered to most dogs.

Drugs used included acepromazine, chlorpromazine, diazepam, midazolam, pentobarbital, phenobarbital, baclofen, methocarbamol, morphine, and butorphanol. Although few comments were made in the medical records regarding the efficacy of specific agents, phenobarbital treatment was not discontinued and replaced by administration of another drug in most instances. Conversely, no dog received methocarbamol for more than 1 day before being switched to a different drug.

Complications—One dog in tetanus severity class III and 8 dogs in class IV had systemic complications during hospitalization. Six dogs developed aspiration pneumonia. Three of these dogs were fed enterally, and 2 received only TPN; nutritional information for 1 dog was unavailable. Four dogs developed gastrointestinal bleeding, and 3 of these dogs had been treated with glucocorticoids before admission to the hospital (prednisone-equivalent doses of 1 mg/kg [0.45 mg/lb], 13 mg/kg [5.9 mg/lb], and 19 mg/kg [8.6 mg/lb]). One dog in class III and 2 dogs in class IV had severe SIRS with multiple organ failure. Of those dogs, 1 had gastrointestinal bleeding, aspiration pneumonia, DIC, and PTE; 1 had gastrointestinal bleeding, DIC, PTE, oliguric renal failure, and acute pancreatitis; and 1 had gastrointestinal bleeding, aspiration pneumonia, DIC, PTE, and myoglobinuria. One of these dogs died, and the other 2 were euthanized; complications in all 3 dogs were confirmed at necropsy.

Survival—Thirty-five dogs were included in the survival analysis, of which 27 were alive at day 28; at that time, 5 dogs had died and 3 dogs had been euthanized. All dogs that died were graded as tetanus severity class IV; 1 dog in class III and 2 dogs in class IV were euthanized. The 28-day survival rate of dogs of higher tetanus severity class was significantly worse (hazard ratio, 6.67; 95% CI, 1.25 to 35.52; P = 0.026) than dogs of lower severity class. Three dogs were lost to followup before 28 days and were censored from all survival analyses. The overall 28-day survival rate for dogs with tetanus was 77%; for dogs that were graded as class III or IV, the 28-day survival rate was 58%.

Clinical effects after 28 days—Of the 27 dogs alive for more than 28 days, information beyond 6 weeks was available for 14 dogs. Ten dogs made a complete recovery, with no clinical signs persisting beyond 60 days that were definitively related to tetanus. Six weeks after having otherwise recovered from tetanus, 2 dogs were reported to have persistent, severe spasms during sleep; the highest tetanus severity class assigned to these dogs was class II and III. The fact that 1 dog in class IV always held 1 thoracic limb in rigid extension during sleep was reported 3 months after its discharge from the hospital. Twelve months after recovery, 1 other dog in class II was reported to persistently cough after eating dry food.

Discussion

Tetanus is uncommon in dogs, likely because of the species' inherent resistance to the tetanospasmin toxin.¹¹ However, results of the present study suggest that once dogs develop severe tetanus, the clinical course of disease is similar to that of severely affected humans.^6,28,29,31 Among 35 dogs included in our study, the 28-day survival rate was 77%, and all dogs in tetanus severity classes I and II survived to 28 days. Fifty percent of the dogs progressed to tetanus severity class III or IV. The high proportion of severely affected dogs in the present study may reflect the patient population at a referral hospital and likely overestimates the prevalence of severe signs in dogs with tetanus.

The timing of appropriate diagnosis and initiation of directed and supportive treatments are important in tetanus patients.²⁹ Because tetanus is uncommon in dogs, clinicians may not include it on their list of differential diagnoses when dogs are in the early stages of the disease and have mild clinical signs. Lack of early recognition leads to delayed diagnosis and treatment. Most of the dogs in the present study had vague clinical signs (as detected by their owners) initially. In 17 of 38 (45%) of dogs, owners first noted enophthalmos, squinting or swollen eyelids, or a generally abnormal appearance to the eyes. Therefore, early-stage tetanus should be a differential diagnosis for mild ocular and facial changes in dogs. Most dogs progressed to a higher tetanus severity classification from the grading that was based on the initial findings of veterinarians. Seven of 17 dogs that were apparently normal or classified as tetanus severity class I on the basis of the initial physical examination findings progressed to class III or IV; thus, even dogs that had mild clinical signs initially may develop severe tetanus. The 28-day survival rate of dogs that were graded as class III or IV was 58%; some of these dogs were managed in an intensive care unit, and others were not. The retrospective nature of the present study prevents assessment of the value of intensive care management of dogs with severe tetanus, but intensive care management improves survival in humans with severe tetanus.^6,10 Dogs graded as tetanus severity class III or IV need continuous vital sign monitoring, fluid therapy, nutritional support, sedatives, and intensive nursing care, all with minimal stimulation.

Autonomic dysfunction has been identified in humans with tetanus and occurs in as many as 30% of humans with the disease.^6,10 The exact mechanism of autonomic syndrome in humans with tetanus is unknown, but the most widely accepted theory involves impairment of inhibitory circuits in the autonomic nervous system.³² Signs of both sympathetic and parasympathetic dysregulation may develop, even in the same patient, resulting in violent autonomic storms.^3,5,6 An alternation between sympathetic and parasympathetic signs was evident in a few class IV dogs in our study, but it was impossible to assess whether these autonomic changes were analogous to the storms detected in affected humans. In humans, autonomic disturbance usually manifests as sympathetic overactivity with sporadic parasympathetic signs, whereas the dogs of this report had more parasympathetic overactivity and fewer sympathetic signs. There have been reports,^4,7 however, of humans with only parasympathetic disturbance, some of whom needed temporary cardiac pacing for severe bradyarrhythmia, as did 1 dog of the present study. Bradycardia and bradyarrhythmias including atrioventricular block and sinus arrest in dogs with tetanus have been previously reported.^15,33 Experimental studies^34,35 in dogs have revealed that sympathetic activation and wide fluctuations in blood pressure develop after injection of tetanus toxin. It is unknown why dogs with naturally occurring tetanus appear to have more parasympathetic overactivity, whereas those receiving experimental injections of tetanospasmin have cardiovascular alterations that more closely mirror humans with autonomic syndrome. The most heavily sedated dogs may have had more autonomic nervous system abnormalities, but the retrospective nature of our study prevents more specific conclusions. For 2 dogs in tetanus severity class IV with SIRS, that syndrome may have contributed to the episodes of tachycardia identified in one and episodes of hypotension identified in the other.

In the present study, dogs with higher tetanus severity class scores were significantly more likely to die than those with less severe signs. Although it seems intuitive that dogs with more severe illness are more likely to die than those less severely affected, it is possible that autonomic nervous system involvement may be an underlying cause of higher mortality rate. Autonomic disturbance is associated with a worse prognosis in humans with tetanus, independent of the severity of neuromuscular signs.³⁶ Seven of the 8 dogs that died or were euthanized had heart rate or blood pressure abnormalities. It is unknown whether the autonomic nervous system abnormalities in these dogs can be attributed solely or in part to tetanospasmin, but whatever the reason for the autonomic signs in this group of class IV dogs, their outcome was significantly worse than dogs without autonomic signs.

Younger dogs were significantly more likely to have a severe clinical course. A similar relationship has been identified in humans, in whom tetanus results in death most frequently among the young and elderly.^8,9 We did not detect this bimodal relationship in the dogs of the present study, and in fact, the median age of 4 years suggests that older dogs may be less likely to develop tetanus than younger dogs. Whether this age bias is because of the small sample size, greater client acceptance of referral for younger dogs, canine behavior, or development of resistance to tetanus toxin over time is unknown. Because naturally occurring antibodies against C tetani toxin have been detected in dogs,³⁷ it is possible that small, subclinical exposures to tetanospasmin can lead to natural immunization.

Comparison of disease course and outcome on the basis of clinical treatment choice is not possible in our retrospective study; such analysis would result in confounding by intention. In other words, more severely affected dogs were more likely to have received these treatments than those dogs that were less severely affected. Therefore, the authors hypothesized that earlier antitoxin administration, appropriate antimicrobial treatment, or wound management would result in improved clinical course and 28-day survival rate, compared with the same treatments administered later. Our study failed to reveal an effect of timing of these treatments on clinical course or outcome in dogs with tetanus. The absence of a significant effect was likely attributable to a type I error associated with small sample size; no effect was identified, although one likely exists. Common sense and human medical literature support early, directed treatment for this disease.²⁹ Results of the present study call into question the usefulness of intradermal skin tests performed prior to antitoxin administration. Two dogs with positive skin test results that received pretreatment with either diphenhydramine or diphenhydramine and dexamethasone had no clinical reaction to antitoxin treatment, whereas 3 dogs with negative skin test results and diphenhydramine pretreatment had reactions. Prospective, controlled studies would be required to determine the importance of intradermal skin testing performed prior to antitoxin administration. Penicillin G or metronidazole is considered an appropriate antimicrobial choice for treatment of C tetani infection.^11,38 However, a large clinical trial revealed that metronidazole is more effective than penicillin G in the treatment of humans with tetanus.³⁹ Therefore, metronidazole should be considered the drug of choice to eradicate C tetani organisms in dogs with tetanus.

The median interval between known wound occurrence and the onset of clinical signs was 9 days (range, 3 to 18 days) in dogs of the present study. This interval range is consistent with previous reports.^11,38 Dogs with surgical introduction of C tetani organisms appeared to be more likely (albeit not significantly) to progress to a higher tetanus severity class than those with external wounds. Internal wounds may have closer proximity to neurons for toxin adherence and spread, compared with wounds in more superficial tissues. The development of tetanus following ovariohysterectomy^40,41 or castration⁴² has been reported previously in dogs. Clinicians should be aware of the possibility of C tetani infection from surgical contamination and should practice strict aseptic surgical technique. In the present study, 6 of 22 (27%) external wounds in the dogs were definitively associated with grass awns and 11 (50%) were draining tracts of the trunk and feet (common following grass awn migration). Clinicians in areas where grass awns are present should be aware that although this plant material is more commonly associated with Actinomyces spp,⁴³ it may also introduce C tetani. In 10 dogs in our study, owners or veterinarians were unable to detect a wound; therefore, the absence of a visible wound does not preclude the diagnosis of tetanus.

Diagnosis of tetanus is based on history and clinical signs. Clostridium tetani was not cultured from wound samples collected from 5 dogs in the present study. These findings are consistent with the human medical literature^9,28,29,31 and do not limit the validity of our study. However, most of the dogs did not undergo diagnostic testing to rule out strychnine toxicosis or generalized myopathies. The authors suspect that the attending clinicians did not perform such tests because the physical examination findings and progression of clinical signs among dogs were highly suggestive of tetanus. It is therefore impossible to ensure retrospectively that all the dogs in our study were infected with C tetani.

Because several sedative drugs were administered to most dogs and few comments were recorded regarding any individual agent's efficacy, it is difficult to draw specific conclusions regarding efficacy of these drugs. Methocarbamol treatment was not given to any dog for more than 1 day before being replaced with a different regimen; this may suggest that it is an ineffective muscle relaxant in the treatment of tetanus. Possibly, inappropriate doses were used because the pharmacokinetics of methocarbamol in dogs are not well understood. The mechanism of action of methocarbamol in humans is thought to involve generalized CNS depression. If this drug works by a similar mechanism in dogs, the effects of tetanospasmin on the CNS may override or reduce methocarbamol's depressant effect. Conversely, treatment with phenobarbital may have had positive effects, given that it was never discontinued in favor of administration of another drug.

Three dogs had hiatal hernias. Hiatal hernia is a reported complication of tetanus in dogs.^44–46 The mechanism is unknown, but it is hypothesized that abnormalities in esophageal motility and tone, combined with diaphragmatic spasms, may play a role.⁴⁴ Three dogs had severe hypoventilation or apnea during tetanic spasms. Spasms of the diaphragm and intercostal muscles can cause acute respiratory arrest. Respiratory failure was the leading cause of death in humans with tetanus prior to the introduction of mechanical ventilation.¹⁰ Dogs with severe tetanic spasms may require mechanical ventilation for survival. Six dogs developed aspiration pneumonia, 2 of which were not fed enterally. Dogs with tetanus may be predisposed to aspiration pneumonia because of lateral recumbency, decreased gag reflex, and trismus, which disallows expulsion of material from the mouth. Of the 4 dogs that developed gastrointestinal bleeding, 3 had received small to large doses of glucocorticoids. It is impossible to know whether the glucocorticoids contributed to ulceration of the gastrointestinal tract in these dogs, but such medications are not recommended in the management of tetanus in humans or dogs.^{11,28,29,31,38} Glucocorticoids should not be used for the treatment of tetanus, particularly because tetanus itself has been associated with gastrointestinal tract ulceration in humans.^6,28,47 Systemic inflammatory response syndrome, renal failure, PTE, and multiple organ failure have all been reported^2,6,10,28,48 in humans with severe tetanus. Such complications developed in 3 dogs in the present study. Systemic inflammation can develop in response to a primary infectious cause such as C tetani infection or as a consequence of tetanus complications such as severe hyperthermia, aspiration pneumonia, or bacterial translocation resulting from intestinal barrier failure. Systemic inflammation leads to widespread coagulation and high circulating concentrations of proinflammatory cytokines, which may cause renal failure and multiple organ dysfunction.

Reports of long-term complications in humans after recovery from tetanus are rare.²⁸ Abnormal movements during sleep were reported for 3 dogs in our study weeks to months after recovering from tetanus. Sleep disturbances following tetanus have been reported^15,17 for 2 other dogs. It is possible that these spasms represent permanent damage to inhibitory circuits within the CNS, although the definitive etiology is not known.

As with all retrospective studies, the present investigation has limitations. Because tetanus is a diagnosis made on the basis of physical examination findings and because C tetani is difficult to culture, definitive diagnosis of this disease is difficult to achieve clinically and impossible to achieve by retrospective data assessment. In addition, it is impossible to know prospectively or retrospectively the cause of autonomic disturbance in dogs with tetanus. Therefore, although autonomic nervous system signs are evident in humans and dogs with tetanus, it is difficult to prove for any individual patient that tetanospasmin is the cause of these derangements.

Our data suggest that the prognosis for dogs with tetanus is good, particularly in dogs that do not develop clinical signs associated with tetanus severity class IV. Younger dogs may be more commonly affected and have a more severe clinical course than older dogs. Surgical introduction of C tetani may be associated with more severe disease than introduction of the organism from external wounds in dogs.