Heart Failure in Dogs With Tick Paralysis Caused by the Australian Paralysis Tick, Ixodes Holocyclus

Fiona E. Campbell, BVSc, PhD, MACVSc

Richard B. Atwell, BVSc, PhD, FACVSc

School of Veterinary Science, The University of Queensland, Queensland, Australia

KEY WORDS: Australian paralysis tick, Ixodes holocyclus, tick paralysis, heart failure, dog

ABSTRACT

To evaluate the cardiovascular changes in dogs with tick paralysis, a prospective clinical investigation of client-owned dogs, which were treated (n = 41) or euthanized (n = 5) for naturally occurring tick paralysis, was performed. The protocol for treated dogs included clinical examination and designation of a gait and respiratory score, measurement of systolic blood pressure and several blood parameters, thoracic radiography, and echocardiography. This protocol was performed on each dog at admission to the hospital, 24 hours later, and at discharge from the hospital. Blood pressure was again assessed approximately 12 months later. From dogs that were to be euthanized with tick paralysis, bronchoalveolar lavage was performed to retrieve pulmonary edema fluid. Dogs with tick paralysis had normal systolic blood pressure, radiographic evidence of pulmonary venous congestion and peribronchial fluid infiltration, and a reduction in echocardiographically derived functional indices. Elevations occurred in packed cell volume, noradrenaline, cortisol, the renin:aldosterone ratio and, in severely affected dogs, lactate. Serum markers of cardiomyocyte structural injury were not altered. Dogs that were euthanized with tick paralysis had pulmonary edema fluid retrieved by bronchoalveolar lavage that was low in protein and consistent with a cardiac origin. This study demonstrates that acute left-sided heart failure occurs in dogs with tick paralysis.

INTRODUCTION

Tick paralysis caused by the Australian paralysis tick, Ixodes holocyclus, is characterized by an ascending flaccid paralysis and, if not treated, death can occur within 24 to 48 hours of onset of clinical signs. Previous work on tick paralysis in experimentally infested dogs concluded that death resulted from factors other than respiratory muscle paralysis, and systemic hypertension, believed to be produced by autonomic dysfunction and excess sympathetic stimulation, was thought to be important in the pathophysiology.1 However, as a result of small group numbers and the invasive experimental method, the results of some of this early work have been recently questioned.2 The aim of this study was to document cardiovascular changes in a large group of dogs with naturally occurring tick paralysis.

MATERIALS AND METHODS

Protocol for Dogs Treated for
Tick Paralysis

Forty-one dogs that were presented to a veterinary hospital for treatment of tick paralysis during the spring and summer of 1999 were selected for the study. Criteria for inclusion were as follows: natural infestation with I. holocyclus, no known preexisting clinical conditions, and current heartworm prophylaxis. Dogs weighing less than 6 kg were excluded because collection of blood (30 mL) might have influenced the hemodynamic status of small dogs.

At admission, a clinical examination was performed and dogs were classified by gait score and respiratory score (Table 1). Systolic blood pressure was determined, and a jugular blood sample (30 mL for packed cell volume [PCV], total protein [TP], plasma renin activity [PRA], plasma aldosterone concentration [PAC], noradrenaline, adrenaline, cortisol, lactate, cardiac troponin-T [cTnT], and lactate dehydrogenase isoenzyme-1 [(LDH1]) was collected. Dogs were weighed and treated. Treatment consisted of acepromazine (ACP; 0.03 mg/kg intravenously; Promex 2, Apex Laboratories, St. Marys, Australia) and dexamethasone sodium phosphate (0.5 mg/kg intravenously; Colvasone, Norbrook Laboratories, Rowville, Australia) followed, 30 minutes later by tick antitoxin serum (TAS; 1 mL/kg; Australian Veterinary Serum Laboratories, Lismore, batch AB/2) given intravenously over 10 to15 minutes by slow manual injection. Thoracic radiography, followed by echocardiography (and electrocardiography for a concurrent study, results not shown), was then performed. This protocol was completed within 2 hours of admission.

Dogs were searched 3 times and ticks were manually removed. To kill any unseen ticks, dogs were bathed with a pyrethrin-based insecticide (Fido’s Flea Itch Concentrate; Mavlab, Slacks Creek, Australia). Dogs were caged individually in an air-conditioned ward (20oC) and no fluids were given per mouth or parentally during the first 24 hours of hospitalization. The ward was continually staffed with experienced veterinary nurses and regurgitation/ vomiting was recorded. Some individual dogs received additional therapy (Appendix).

Twenty-four hours after treatment, the protocol was repeated in the following order: clinical examination, measurement of blood pressure, blood collection (2 mL for PCV and TP), radiography, and echocardiography. After completion of the protocol, oral fluids were gradually introduced to dogs that had not regurgitated/vomited within the preceding 12 hours. Except for collection of a blood sample (2 mL), this protocol was not performed on dogs (n = 9) that were clinically normal at 24 hours. In these dogs, oral fluids were introduced and the discharge protocol was performed 12 to 24 hours later.

Dogs were eligible for discharge from the hospital when clinically normal and when drinking water normally ad libitum for as least 12 hours. When these criteria were met, dogs were clinically examined, blood pressure was measured, and blood collected (30 mL for PCV, TP, PRA, PAC, noradrenaline, adrenaline, cortisol, lactate, cTnT, and LDH1). Like at admission, 0.03 mg/kg acepromazine was given intravenously and radiography and echocardiography performed 30 to 40 minutes later.

Approximately 12 months later, a subgroup of these dogs was admitted to the hospital. A clinical examination was performed and systolic blood pressure was again determined.

Systolic Blood Pressure

Blood pressure was measured indirectly using an ultrasonic Doppler (Model 811–AL; Parks Medical Electronics, Aloha, USA). All measurements were performed in a quiet room. The dog was allowed to find a comfortable position in sternal recumbency. Dogs that would not settle were lightly restrained. The circumference of the mid-antebrachial region on a chosen forelimb was measured and an occlusive cuff (Critikon; Johnson & Johnson, NJ, USA) was selected such that its width was approximately 40% of the circumference of the limb. The cuff was secured around the mid-antebrachial region and connected to a sphygmomanometer (Speidel and Keller, Disytest, Jungingen, Germany). The hair proximal to the palmar metacarpal pad was clipped and the pulse of the superficial palmar artery was located by auscultation. Coupling gel (General Imaging, ATL Ultrasound) was applied to the probe that was then fixed in position over the artery with adhesive tape. The cuff was inflated to suprasystolic values and slowly deflated. Systolic pressure was defined as the first crisp note. At least 5 measurements were recorded over a 10-minute period, and values were averaged to obtain the final result. The limb circumference, cuff size, and chosen forelimb were recorded and the same method repeated on subsequent assessments of each dog.

Blood Collection and Analysis

Blood was collected by jugular venipuncture and transferred into edetic acid (EDTA), lithium heparin with 1 to 2 mg of sodium metabisulphate (BDH AnalaR, Poole, UK) and serum-separated vials. Samples were centrifuged for 10 minutes at 1500 g. Serum and plasma were separated from the clot, transferred to cryogenic storage vials, and stored at –20oC within 1 hour of collection.

PCV was measured from whole blood by the microhematocrit method using 40-mm capillary tubes and a microhematocrit centrifuge (StatSpin MP, Norwood, MA). TP was measured from the microhematocrit tube plasma sample using a handheld refractometer (Leica 10436; Reichert Inc, Depew, USA). Both tests were performed in duplicate and recorded as the mean.

PRA was determined from plasma (EDTA) using the GammaCoat [125I] plasma renin activity radioimmunoassay kit (Dia Sorin Inc, Minnesota, MN). PAC was measured from serum using the Coat-A-Count aldosterone radioimmunoassay kit (Diagnostic Products Corporation, Los Angeles, CA). Both tests were performed at Sullivan Nicolaides Pathology (Brisbane, Australia) using the 1277–GammaMaster gamma counter (LKB Wallac, Uppsala, Sweden). The ratio of PRA: PAC was calculated as previously described.3

Noradrenaline and adrenaline were determined at the Division of Chemical Pathology Princess Alexandra Hospital (Brisbane, Australia) from plasma (lithium heparin with sodium metabisulfite) using reverse-phase isocratic and electrochemical detection high-performance liquid chromatography.4

Cortisol was measured from serum samples at the Division of Chemical Pathology Princess Alexandra Hospital using the Bayer Centaur Immunoassay System (Bayer Diagnostics, Leverkusen, Germany). When cortisol was not detectable below 30 nmol/L, a value of 30 nmol/L was assigned for statistical purposes.

Lactate concentration was determined at the Division of Veterinary Pathology and Anatomy, School of Veterinary Science (University of Queensland, Australia) from plasma (EDTA) using the lactate MPR1 kit (Boehringer Mannheim, Indianapolis, IN) and Roche Cobas Mira (Roche Diagnostic Systems, Indianapolis, IN).

Serum was analyzed for cTnT at Sullivan Nicolaides Pathology using the Troponin–T immunoassay and Elecsys 2010 (Roche Diagnostic Systems).

Electrophoretic separation of lactate dehydrogenase isoenzymes was performed at Sullivan Nicolaides Pathology using a commercial kit (Paragon, Beckman, Fullerton, CA). The volume of the LDH1 bands was determined using computer analysis (ImageQuaNT, version 1.2, Molecular Dynamics, San Francisco, CA).

Radiography

Radiography (Seimans Heliophos 4M) was performed using 400 speed rare earth screens (Fuji HR-G30) with an automatic developer (Agfa-Gevaert). Left lateral and dorsoventral projections were taken at full inspiration, when possible. Exposure factors were recorded for each dog and repeated for serial radiographs.

On the lateral radiograph, the diameters of the right cranial lobar artery and vein were measured and compared with the width of the right fourth rib as previously described.5 Similarly, on the dorsoventral projection, the diameters of the right and left caudal lobar artery and veins were measured and compared with the width of the ninth rib.6

On the dorsoventral radiograph, the outer and luminal diameters of the right middle lobe bronchus were measured to determine the thickness of peribronchial cuffing, a reflection of interstitial infiltrate. The width of the right eighth rib at its emergence from the spine was measured for comparison.

To facilitate objective comparison of serial radiographs, artery and vein diameters and bronchial wall thickness were divided by their corresponding rib widths. Likewise, arterial diameters were divided by the diameter of the paired vein. The values obtained for the right cranial lobar artery and vein and right and left caudal lobar arteries and veins were averaged to produce an artery:rib, vein:rib, and artery:vein ratio for each dog at admission, at 24 hours, and at discharge.

Echocardiography

The right thorax was clipped and the dog was lightly restrained in right lateral recumbency on a table designed to allow access to the underside of the animal. A lead II electrocardiogram (ECG) trace was simultaneously recorded during echocardiography and displayed on the ultrasound monitor. The 3 ECG electrodes were attached to the standard recording positions immediately proximal to the elbows and stifle.7 Alcohol was applied at the point of lead attachment to enhance electrical conductivity.

Cardiac ultrasound (Ausonics Impact VFI; Lane Cove, Australia) was performed with the transducer (5-mHz linear array, Ausonics) positioned underneath the animal and the beam directed upward through the right parasternal region at the point of the strongest palpable beat. Ultrasound transmission gel (General Imaging, ALT Ultrasound) was used to ensure airtight coupling between the transducer and the thorax.

A short axis view of the left ventricle at the level of the papillary muscles was obtained. M-mode images were produced by directing the scanner to bisect the left ventricular lumen through the 2 papillary muscles. Measurements were taken in accordance with the guidelines of the American Society of Echocardiography8 using the leading-edge method. The calipers were positioned at the beginning of the QRS complex for measurement of the left ventricular internal diameter at end-diastole (LVIDd). The left ventricular internal diameter at systole (LVIDs) was measured at the nadir of systolic septal motion independent of the ECG complex.

To calculate fractional shortening (FS), left ventricular end-diastolic volume (LVEdV), left ventricular end-systolic volume (LVEsV), ejection fraction (EF), and cardiac output (CO), the following formula were used:

FS (%) = (LVIDd – LVIDs)/(LVIDd) ¥ 100

LVEdV (mL) = 7.0 ¥ (LVIDd)3/(2.4 + LVIDd)

LVEsV (mL) = 7.0 ¥ (LVIDs)3/(2.4 + LVIDs)

EF (%) = (LVEdV – LVEsV)/LVEdV ¥ 100

CO (mL/min) = Stroke volume (SV) ¥ heart rate (HR)

Where SV (mL) = LVEdV – LVEsV

Values were obtained from at least 3 cardiac cycles and averaged to give the final values.

Effect of TAS

A concurrent study was performed to identify any effect that TAS might have on echocardiographic values. Echocardiography was performed on healthy pound-sourced dogs (n = 6) of various breeds, both genders, and unknown ages. TAS (1 mL/kg) was given as described previously and echocardiographic examination repeated 30 to 40 minutes later.

Protocol for Dogs Euthanized With
Tick Paralysis

Client-owned dogs with tick paralysis were sourced as before. Criteria for inclusion was as described earlier plus the client’s request for euthanasia of the dog with tick paralysis as a result of financial constraints and/or severity of disease.

Blood Collection

The cephalic vein was catheterized and a blood sample (5 mL) was immediately collected and transferred into a serum-separated vial. Samples were centrifuged and the serum separated and stored as before.

Bronchoalveolar Lavage (BAL)

Anesthesia was induced by thiopentone sodium (5% Pentothal; Rhone Merieux, Poulenc, France) given intravenously to effect. Secretions were cleared from the pharynx using a surgical suction device (Electrolux). The dog was intubated with a suitably sized endotracheal tube (Mallinckrodt) and a sterile urinary catheter (Sovereign) was advanced down the endotracheal tube until the tip wedged in a small bronchus.

A 20-mL aliquot of warmed sterile 0.9% saline (Baxter) was infused through the catheter and immediately withdrawn by gentle suction until no further fluid was obtained. This was repeated with a second 20-mL aliquot of saline. The volume of fluid recovered was measured and stored at -20°C in cryogenic storage vials. Pentobarbitone sodium (Lethobarb; Virbac, Peakhurst, Australia) was administered rapidly intravenously to effect euthanasia.

Fluid Analysis and Calculations

TP and urea concentration of serum and of the fluid retrieved by BAL was determined at the Division of Veterinary Pathology and Anatomy using the Unimate-5 urea kit (Roche Diagnostics) and Unimate-7 Total Protein Kit (Roche Diagnostics) with the Roche Cobas Mira (Roche Diagnostics).

The actual volume of edema fluid retrieved by BAL was determined using urea, which freely diffuses throughout the body, as an endogenous dilution marker and the formula9:

Volume of edema fluid (mL) = [urea (mg/ mL) in BAL ¥ volume of BAL retrieved (mL)]/urea (mg/ mL) in serum

The TP concentration in the edema fluid was calculated using the formula:

TP (mg/ mL) in edema fluid = [TP (mg/mL) in BAL ¥ volume of BAL retrieved (mL)]/volume of edema fluid (mL)

The ratio of edema fluid to serum TP concentration was then calculated.

Statistical Methods

Statistical analysis was performed using computer software (Prism, GraphPad).

Changes in each parameter (systolic blood pressure, PCV, TP, echocardiographic values, and radiographic ratios) over the course of the investigation (admission, 24 h, discharge, and 12 mo) were assessed by repeated-measures analysis of variance (ANOVA). Tukey’s test was subsequently performed when significant differences were identified. Paired t-tests (2-way) were applied to compare blood parameters at admission and discharge (PRA, PAC, adrenaline, noradrenaline, cortisol, lactate, and LDH1) and in the concurrent study of healthy dogs to compare echocardiographic values obtained before and after TAS administration.

On single occasions (admission or discharge), 1-way ANOVA tests were applied to identify associations between clinical score (gait or respiratory score) and measured parameter (blood pressure, PCV, TP, adrenaline, noradrenaline, cortisol, and lactate) when data was normally distributed. The Kruskal-Wallis test was used similarly for nonparametric data (renin and aldosterone). Tukey’s test for parametric or Dunn’s test for nonparametric data were subsequently performed when significant differences were identified.

Spearman’s rank correlation was used to test for association between gait and respiratory scores.

Except when specified, values are given as mean ± standard error of mean and include all available data. AP value of P <0.05 was accepted as significant.

RESULTS

Dogs Treated for Tick Paralysis

Forty-one dogs, 20 female and 21 male, aged 6 months to 10 years (mean ± standard deviation [SD], 4.7 ± 3.1 y) and weighing 6 to 50.8 kg (mean ± SD, 25.9 ± 9.4 kg) were studied. Breeds, or observed primary crosses of dogs, studied included: Border Collies (10), German Shepherds (7), Australian Cattle Dogs (4), Labrador Retrievers (4), Staffordshire Bull Terriers (4), Australian Kelpies (2), Bull Mastiffs (2), Boxer (1), Doberman (1), English Springer Spaniel (1), German Shorthaired Pointer (1), Greyhound (1), Poodle (1), Rottweiler (1), and Siberian Husky (1).

Gait and Respiratory Scores

Respiratory and gait scores (Table 1A and B) were correlated in individual dogs (P = 0.0006).

Systolic Blood Pressure

The mean systolic blood pressure was similar on all occasions: 161 ± 7 mm Hg at admission, 166 ± 9 mm Hg at 24 hours, 180 ± 8 mm Hg at discharge, and 160 ± 12 mm Hg at 12 months (Fig. 1). Systolic blood pressure at admission was not associated with respiratory or gait scores.

Blood Analysis

The mean PCV was higher at admission than 24 hours and decreased further as dogs recovered (P <0.0001) (Table 2). There was no association between PCV at admission and gait score. However, there was a relationship between PCV and respiratory score (P = 0.01); dogs with respiratory score D (59 ± 1.5%; n = 3) had a higher PCV than dogs with respiratory score B (48 ± 1.1%; n = 19). The mean TP at admission was lower than mean TP at discharge (P = 0.01) (Table 2). There was no association between TP at admission and gait or respiratory scores.

The mean PRA and PAC were lower at admission than at discharge (Table 3). There was no association between PRA at admission and gait or respiratory scores. Likewise, no association was identified between PAC at admission and gait or respiratory scores. The PRA:PAC ratio was higher at admission than at discharge (Table 3). There was no association between the PRA:PAC ratio (at admission or discharge) and gait score or respiratory scores.

The mean noradrenaline concentration at admission exceeded the mean noradrenaline concentration at discharge (Table 3). There was no association between noradrenaline concentration at admission and gait score. However, noradrenaline concentration was associated with respiratory score (P = 0.02); dogs with respiratory score D (8.50 ± 5.34 nmol/L; n = 3) had a higher noradrenaline concentration than those with respiratory score B (1.53 ± 0.29 nmol/L; n = 12). Adrenaline concentration was similar at admission and discharge (Table 3) and was not associated with gait or respiratory scores.

There was no relationship between cortisol concentration at admission and gait or respiratory score. At discharge, only 2 dogs had detectable cortisol concentrations (51 nmol/L and 38 nmol/L). Accordingly, cortisol concentration was significantly lower at discharge that at admission (Table 3).

Mean lactate concentration at admission and discharge was similar (Table 3). There was no association between lactate concentration at admission and gait score. However, there was a link between lactate and respiratory score (P = 0.045); dogs with respiratory score D (3.70 ± 1.97 mmol/L; n = 3) had a higher lactate concentration than dogs with respiratory score C (1.69 ± 0.10 mmol/L; n = 9).

CTnT was not detectable in any sample at either admission or discharge, and the volume of the LDH1 electrophoretic bands was similar at admission and discharge (Table 3).

Radiography

The mean artery:rib ratio determined from radiographs taken at admission, 24 hours, and discharge did not vary significantly. Likewise, the reducing trend of the mean vein:rib ratio over the course of recovery was not significant. However, the mean artery:vein ratio at admission was 0.93 ± 0.02 and was significantly lower than the artery:vein ratio of 0.98 ± 0.02 determined from radiographs taken at discharge (P = 0.048) (Fig. 2).

The peribronchial cuff:rib ratio decreased significantly as dogs with tick paralysis recovered (P = 0.01) (Fig. 3).

Echocardiography

The diastolic chamber dimensions (LVIDd and LVEdV) increased significantly (P <0.0001) as dogs recovered (Table 4). Likewise, significant increases were identified from admission to discharge for FS (P <0.0001), EF (P <0.0001), and CO (P = 0.04). The mean heart rate recorded at echocardiographic examination was higher at admission than at discharge (P <0.0001) and heart rate was lower on both occasions than heart rate determined by auscultation before ACP and echocardiography (P = 0.0002 and P = 0.002, respectively) (Table 4).

Effect of TAS

TAS administered to healthy dogs in a concurrent study did not produce any alteration in diastolic or systolic chamber dimensions, FS, EF, or CO. This indicates that cardiac mechanical function is not changed by TAS and validates the direct comparison of echocardiographic values obtained from dogs with tick paralysis at admission, 24 hours, and discharge.

Dogs Euthanized for Tick Paralysis

The mean edema fluid to serum protein ratio of dogs euthanized for tick paralysis was 0.18 ± 0.08 (n = 5) (Table 5).

Discussion

Systolic Blood Pressure

The mean systolic blood pressure of dogs was similar at admission with tick paralysis, 24 hours after admission, at discharge when clinically normal, and when reassessed 12 months later. This indicates that dogs in the current investigation did not develop marked changes in systolic blood pressure and suggests that, in contradiction to one earlier experimental study,1 systolic arterial hypertension does not occur with tick paralysis.

Interestingly, when systolic blood pressure of dogs at admission and discharge is compared using paired t-tests, blood pressure is significantly higher in dogs at discharge than at admission (P = 0.02; n = 35), perhaps as a result of the stress of hospitalization.10

Blood Analyses

The elevated PCV of dogs in the present study concurs with earlier observations of dogs with tick paralysis,1 which were thought to have a high PCV as a result of dehydration or splenic contraction. However, serial measurements in this investigation showed that PCV fell significantly within 24 hours of admission. Because dogs had no access to fluids during this time, the higher PCV at admission cannot be attributed to dehydration. Likewise, splenic contraction in dogs exposed to various stresses and exercise produces less than an 8% increase in PCV11,12 such that the 16% increase in PCV of dogs at admission relative to discharge might not be solely attributed to splenic contraction.

The high PCV of dogs with tick paralysis could result in part from fluid shifting from the vasculature into the pulmonary interstitium and alveoli as pulmonary edema develops. This is supported by the reduction in PCV at 24 hours, which might occur as edema fluid is resorbed back into the vasculature as cardiovascular function improves, and by the association between PCV and respiratory score in tick-paralyzed dogs that could occur because a large fluid shift would concurrently produce severe pulmonary edema and high PCV.

The mean PRA and PAC of dogs at admission with tick paralysis is comparable with values obtained from normal dogs.13-15 This might be explained if at the first sampling time, the RAAS had not yet responded to the acute cardiovascular changes. Experimental reduction in renal perfusion pressure takes some time to activate the RAAS,16 and the rise in PRA and PAC following experimental constriction of the caudal vena cava of dogs takes up to 24 hours to peak.17 The PRA:PAC ratio of dogs with tick paralysis also supports early activation of the RAAS. The mean PRA:PAC ratio of 21 ± 3 hours-1 in dogs admitted with tick paralysis seems higher than the PRA:PAC ratio of normal dogs, which is typically less than 8 hours-1.3 Furthermore, the PRA:PAC ratio of dogs in this study was significantly greater at admission than at discharge. This reflects an increase in PRA relative to PAC in dogs with tick paralysis and typifies early activation of the RAAS in which PRA secretion precedes the release of PAC.18

An elevation in noradrenaline in healthy dogs exposed to stressful stimuli is accompanied by similar increases in adrenaline.12,19 However, dogs and people with heart failure have increased serum noradrenaline without elevations in adrenaline.20,21 Furthermore, in dogs and people with congestive heart failure, noradrenaline concentration is positively correlated with clinical severity and is a significant predictor of mortality.20,22 The elevation of noradrenaline, without an accompanying increase in adrenaline, and the positive association between respiratory score and noradrenaline concentration suggests that, in dogs with tick paralysis, the sympathetic nervous system is predominantly activated by cardiac dysfunction rather than by stress.

The mean cortisol concentration of dogs with tick paralysis is comparable with the cortisol concentration of dogs exposed to physiological stresses such as housing, noise,19 transportation,11 surgery, and non-adrenal illness.23 Dexamethasone administered at single intravenous doses 5-fold lower than used in the current study reduced plasma cortisol of healthy dogs for up to 32 hours,24 and because the duration of suppression of the hypothalamic-pituitary-adrenal axis by dexamethasone is dose-dependent,25 the reduction in cortisol concentration of dogs at discharge might be attributable to the dexamethasone administered at admission.

Blood lactate concentration rises when reduced tissue oxygenation causes cellular metabolism to shift from aerobic to anaerobic,26 and it is used to assess severity of shock in ill and injured dogs and to evaluate cardiac function in dogs with heart failure.27 In the present investigation, dogs with a respiratory score D had a significantly higher lactate concentration than other dogs with tick paralysis. Reduced tissue oxygenation in these dogs could result from reduced cardiac output as a result of cardiac dysfunction and/or reduced arterial oxygenation arising from pulmonary congestion and edema.

CTnT is a myofibrillar protein found predominantly in cardiac myocytes, and its presence in serum is a highly sensitive marker of cardiac injury. Nondetectable cTnT concentration is typical of normal dogs, and in these dogs with tick paralysis, it reflects the lack of myocardial cell damage.28

In the current study, LDH1, the isoenzyme found predominantly in heart muscle,29 did not differ in the serum of dogs between admission and discharge. LDH1 is released slowly following cardiac insult and persists in serum for several days,30 so an increase in LDH1 activity at discharge would be expected if significant cardiac injury occurred, and the lack of change in LDH1 levels indicates that myocyte structural damage does not occur with tick paralysis.

Radiography

On all occasions in the current study, the mean artery:rib ratio and vein:rib ratio were less than one, indicating that the size of the pulmonary vessels were within the limits of normal.5,6 The mean artery:vein ratio at discharge was 0.98 ± 0.02, indicating that pulmonary arteries and veins were effectively equal in size. However, this ratio was significantly lower in dogs at admission, reflecting the relative enlargement of the pulmonary veins in dogs with tick paralysis (Fig. 2). Pulmonary venous congestion occurs as pulmonary venous pressure increases in response to left ventricular dysfunction, and it is an early and highly significant sign of cardiogenic pulmonary edema.6,31

On radiographs of dogs at admission with tick paralysis, the peribronchial cuff thickness of the right middle lobe bronchus was significantly greater than at discharge (Fig. 3). Peribronchial cuff formation occurs as cardiac failure progresses, and increasing hydrostatic pressure causes transudation of fluid from the pulmonary capillary bed into the interstitium.6,31 The reduction in the thickness of the peribronchial cuff with recovery from tick paralysis is an indication of resolution of interstitial pulmonary edema; however, peribronchial cuff thickness is not a measure of the severity of edema because the degree of fluid accumulation around one bronchus does not reflect overall lung involvement. It discounts the hilar distribution of cardiogenic pulmonary edema, heterogenous lymphatic drainage, and hydrostatic fluid accumulation associated with recumbency31; and the radiographic signs of acute pulmonary edema lag slightly behind observed clinical signs so that radiographs do not directly reflect the current severity of pulmonary edema.6

Echocardiography

FS and EF were reduced in dogs with tick paralysis largely as a result of a reduction in mean diastolic left ventricular chamber dimension (Table 4). At admission, any reduction in preload resulting from a fluid shift from the vasculature into the pulmonary interstitium might have contributed to the reduction in diastolic left ventricular chamber dimension and FS. However, one study of cats showed that when preload was reduced by removal of 25% of blood volume, the FS fell by less than 10%.32 Likewise, the reduction in preload reflected by a 17% increase in PCV after exercise in a group of horses caused a 9% fall in FS.33 Similar studies in dogs are lacking, but in these dogs with tick paralysis, comparable increases in PCV accompanied a 35% reduction in FS suggesting that the reduction in FS cannot be attributed solely to altered preload.

Altered loading conditions have also been associated with the increase in blood viscosity that accompanies high PCV.34 However, the mean PCV of dogs at admission with tick paralysis, although significantly elevated, remained within the limits of the normal range, and changes in blood viscosity and secondary cardiovascular effects are therefore unlikely to be appreciable.

The elevated heart rate of dogs with tick paralysis at admission might have reduced diastolic left ventricular chamber dimension and FS by limiting ventricular filling time.35 However, the changes in heart rate in dogs with tick paralysis were modest, and more recent studies indicate that diastolic and systolic chamber dimensions are both similarly reduced at high heart rates such that no actual change in FS is produced.36

At 24 hours, mean FS was higher than at admission but remained significantly lower than the value at discharge. PCV had fallen significantly, limiting any effects of preload on the reduced mean FS at 24 hours. Systolic blood pressure was unchanged and heart rate had returned to normal, which tended to exclude any effects of afterload and heart rate on FS. Therefore, at 24 hours, the reduction in FS was predominantly the result of intrinsic cardiac dysfunction. Although it is impossible to determine from these results the degree to which loading conditions and heart rate affected FS at admission, a primary alteration in myocardial function is most likely to have been the major determinant of reduced FS at that time also.

This reduction in diastolic chamber dimension and FS in dogs with tick paralysis could be the result of toxin-induced myocardial dysfunction of a predominantly diastolic nature. This is supported by in vitro studies on isolated rat ventricular papillary muscles that identified that the toxin of I. holocyclus specifically prolongs myocardial relaxation time.37

The systolic chamber dimensions of the left ventricle were unchanged with tick paralysis, indicating that systolic function and the contractile state of the myocardium was maintained. This concurs with earlier work in which myocardial contractility assessed through cardiac catheterization of experimentally infested dogs was unchanged by tick paralysis.1 In other forms of diastolic dysfunction, such as hypertrophic heart disease, the myocardium compensates during systole for diastolic myocardial dysfunction to preserve FS.38 Lack of systolic compensation in tick paralysis might be a consequence of the acute nature of diastolic compromise or, in the presence of sympathetic activation, could indicate that contractile reserve is limited or myocardial responsiveness is retarded.

Bronchoalveolar Lavage and Edema Fluid Assessment

Studies in human patients have determined that when pulmonary edema arises from cardiac dysfunction, the protein concentration in the edema fluid is less than half that in blood, whereas, when the cause of edema is noncardiac, the protein ratio of edema fluid to serum is greater than 0.7.39 In the current investigation, all dogs had a concentration of protein in pulmonary edema fluid that was less than half the serum protein concentration, which is consistent with a cardiac cause and is supportive evidence that left-sided heart failure produces pulmonary edema in dogs with tick paralysis.

Low protein concentration in the edema fluid retrieved from 2 dogs resulted in edema to serum protein ratios of 0.07 or less. This could be explained if normal epithelial lining fluid, rather than edema fluid, was retrieved by BAL.40 This could occur because general alveolar flooding had not yet developed or because the distribution of pulmonary edema was heterogenous; and without the aid of selective bronchoscopy, it was not possible to lavage an area of the lung where the alveoli were known to be fluid-filled.

There are a number of limitations to this clinical investigation. First, pulmonary vascular pressures and CO were not directly measured because the use of client-owned dogs precluded pulmonary artery catheterization, a technique which has been associated with increased morbidity and mortality of human patients.41 Secondly, FS and EF are crude indices that use ventricular measurements at 2 single selected time points in the cardiac cycle to represent the whole dynamic contractile process, and tissue Doppler imaging is indicated to better assess the proposed diastolic dysfunction in dogs with tick paralysis. Third, few dogs were available for BAL and more samples, collected with the aid of bronchoscopy, are necessary to statistically validate the low protein nature of the pulmonary edema fluid. Finally, the low numbers of dogs studied with severe tick paralysis might be accountable for the inability to correlate some parameters with clinical score.

In summary, this study has identified cardiovascular changes in dogs with tick paralysis, which are consistent with left-sided congestive heart failure. These include pulmonary venous congestion and interstitial infiltrate, a reduction in FS, largely attributable to diastolic dysfunction, and low protein pulmonary edema fluid. Changes in PCV reflect a fluid shift from the vasculature to the interstitium as pulmonary edema develops, and changes in PRA and PAC are indicative of early activation of the RAAS in response to reduced cardiac output. Alterations in catecholamines reflect sympathetic activation in heart failure, and an elevation in blood lactate, in dogs severely affected by tick paralysis, is evidence of tissue hypoxia. Serum LDH1 and cTnT indicate that this altered cardiac function is not accompanied by structural injury.

This study has implications for treatment of dogs with tick paralysis, which must include therapeutics for acute heart failure such as oxygen therapy, diuretics, and vasodilators. Use of calcium channel-blocking or beta-adrenoceptor-blocking drugs could be indicated if further studies confirm a predominantly diastolic nature of heart failure.

Appendix

Additional therapeutics were administered to dogs as follows: furosemide (Frusemide; Troy Laboratories, Smithfield, Australia) 5 mg/kg intravenous single dose after completion of admission protocol; n = 3), metoclopramide (Metomide; Delvet, Seven Hills, Australia; 0.5 mg/kg intramuscularly or intravenously 3 times per day up to 5 doses; n = 12), amoxicillin/clavulanic acid (Clavulox; Pfizer, West Ryde, Australia; 12 mg/kg orally twice per day for 5 days; n = 3), parenteral fluids (compound sodium lactate, Baxter; 2 mL/kg per hour intravenously after completion of 24-hour protocol continued until oral fluids could be introduced; n = 5).

Additional therapeutic procedures were performed as follows: bladder catheterization (n = 2), surgical suction of pharynx (n = 6), and flow-by oxygen (n = 2).

Acknowledgments

The authors thank Christine Kidd and Pauline Gaven of Manly Road Veterinary Hospital, Brisbane, for allowing the study to be performed using their hospital facilities and admitted cases of tick paralysis. Financial support was provided by the Canine Control Council and the Australian Companion Animal Health Foundation and is gratefully acknowledged.

REFERENCES

1. Ilkiw JE: A Study of the Effects in the Dog of Ixodes holocyclus [PhD thesis]. University of Sydney, New South Wales; 1979.

2. Atwell R, Fitzgerald M: Unsolved issues in tick paralysis. Aust Vet Practit 24:155-161, 1994.

3. Pedersen HD, Olsen LH, Arnordottir H: Breed differences in the plasma renin activity and plasma aldosterone concentration of dogs. J Vet Med Assoc 42:435-441, 1995.

4. Engeland WC, Bereiter DF, Gann DS: Sympathetic control of adrenal secretion of enkephalins after hemorrhage in awake dogs. Am J Physiol 251:R341-R348, 1986.

5. Thrall DE, Losonsky JM: A method of evaluating canine pulmonary circulatory dynamics from survey radiographs. J Am Anim Hosp Assoc 12:457-462, 1976.

6. Thrall DE: Textbook of Veterinary Diagnostic Radiology. Philadelphia: Saunders; 1998.

7. Detweiler DK: The dog electrocardiogram: A critical review. In: MacFarlane PW, Lawrie TDV, eds. Comprehensive Electrocardiography: Theory and Practice in Health and Disease. New York: Pergamon Press; 1988:1268-1329.

8. Sahn DJ, DeMaria A, Kisslo J, Weyman A: Recommendations regarding quantitation in M-mode echocardiography: Results of a survey of echocardiographic measurements. Circulation 58:1072-1083, 1978.

9. Rennard SI, Basset G, Lecossier D, et al: Estimation of volume of epithelial lining fluid recovered by lavage using urea as a marker of dilution. J Appl Physiol 60:533-538, 1986.

10. Coulter DB, J.C. K. Blood pressures obtained by indirect measurements in conscious dogs. J Am Vet Med Assoc 184:1375-1378, 1984.

11. Kuhn G, Lichtwald K, Hardegg W, Abel HH: The effect of transportation stress on circulating corticosteroids, enzyme activities and hematological values in laboratory dogs. J Exp Anim Sci 34 99-104, 1991.

12. Joles JA, den Hertog JM, Huisman GH, et al: Plasma renin activity and plasma catecholamines in intact and splenectomized running and swimming Beagle dogs. Eur J Appl Physiol 49:111-119, 1982.

13. Koch J, Pedersen HD, Jensen AL, Flagstad A: Activation of the renin-angiotensin system in dogs with asymptomatic and symptomatic dilated cardiomyopathy. Res Vet Sci 59:172-175, 1995.

14. Knowlen GG, Kittleson MD, Nachreiner RF, Eyster GE: Comparison of plasma aldosterone concentration among clinical status groups of dogs with chronic heart failure. J Am Vet Med Assoc 183:991-996, 1983.

15. Pedersen HD, Koch J, Poulsen K, Jensen AL, Flagstad A: Activation of the renin-angiotensin system in dogs with asymptomatic and mildly symptomatic mitral valvular insufficiency. J Vet Int Med 9:328-331, 1995.

16. Cowley AW, Miller JP, Guyton AC: Open-loop analysis of the renin-angiotensin system in the dog. Circulation Res 28:568-581, 1971.

17. Watkins L, Burton JA, Haber E, et al: The renin-angiotensin-aldosterone system in congestive failure in conscious dogs. J Clin Invest 57:1606-1617, 1976.

18. Vieweg WVR, Veldhuis JD, Carey RM: Temporal pattern of renin and aldosterone secretion in men: Effects of sodium balance. Am J Physiol 262:F871-F877, 1992.

19. Engeland WC, Miller P, Gann DS: Pituitary-adrenal and adrenomedullary responses to noise in awake dogs. Am J Physiol 27:R672-R677, 1990.

20. Ware WA, Lund DD, Subieta AR, Schmid PG: Sympathetic activation in dogs with congestive heart failure caused by chronic mitral valve disease and dilated cardiomyopathy. J Am Vet Med Assoc 197:1475-1481, 1990.

21. Francis GS: Neurohumoral mechanisms involved in congestive heart failure. Am J Cardiol 55:15A-21A, 1985.

22. Cohn JN, Levine TB, Olivari MT, et al: Plasma norepinephrine as a guide to prognosis in patients with chronic congestive heart failure. N Engl J Med 311:819-823, 1984.

23. Church DB, Nicholson AI, Ilkiw JE, Emslie DR: Effect of non-adrenal illness, anaesthesia and surgery on plasma cortisol concentrations in dogs. Res Vet Sci 56:129-131, 1994.

24. Kemppainen RJ: Effects of single intravenous doses of dexamethasone on baseline cortisol concentrations and responses to synthetic ACTH in healthy dogs. Am J Vet Res 45:742-746, 1984.

25. Meikle AW, Tyler FH: Potency and duration of action of glucocorticoids. Am J Med 63, 1977.

26. Lagutchik MS, Ogilvie GK, Wingfield WE, Hackett TB: Lactate kinetics in veterinary critical care: A review. J Vet Emerg Crit Care 6:81-95, 1996.

27. Kittleson MD, Johnson LE, Pion PD: Submaximal exercise testing using lactate threshold and venous oxygen tension as endpoints in normal dogs and in dogs with heart failure. J Vet Int Med 10:21-27, 1996.

28. O’Brien PJ, Dameron GW, Beck ML: Differential reactivity of cardiac and skeletal muscle from various species in two generations of cardiac troponin-T immunoassays. Res Vet Sci 65:135-137, 1998.

29. Wroblewski F, Gregory KF: Lactic dehydrogenase isozymes and their distribution in normal tissues and plasma and in disease states. Ann New York Acad Sci 94:912-931, 1961.

30. Usategui-Gomez M, Wicks RW, Warshaw M: Immunochemical determination of the heart isoenzyme of lactate dehydrogenase (LDH1) in human serum. Clin Chem 25:729-734, 1979.

31. Suter PF, Lord PF: Thoracic Radiography: A Text Atlas of Thoracic Diseases of the Dog and Cat. Wettswill, Switzerland; 1984.

32. Moise NS, Horne WA, Flanders JA, Strickland D: Repeatability of the M-mode echocardiogram and the effects of acute changes in heart rate, cardiac contractility, and preload in healthy cats sedated with ketamine hydrochloride and acepromazine. Cornell Vet 76:241-258, 1986.

33. Bertone JJ, Paul KS, Wingfield W, Boone JA: M-mode echocardiographs of endurance horses in the recovery phase of long-distance competition. Am J Vet Res 48:1708-1712, 1987.

34. McGrath RL, Weil JV: Adverse effects of normovolemic polycythemia and hypoxia on hemodynamics in the dog. Circulation Res 43:793-798, 1978.

35. Bristow JD, Ferguson RE, Mintz F, Rapaport E: The influence of heart rate on left ventricular volume in dogs. J Clin Invest 42:649-655, 1963.

36. Jacobs G, Mahjoob K: Influence of alterations in heart rate on echocardiographic measurements in the dog. Am J Vet Res 49:548-552, 1988.

37. Campbell FE. The Cardiovascular Effects of the Toxin(s) of the Australian Paralysis Tick, Ixodes holocyclus [PhD thesis]. University of Queensland, Brisbane, Queensland; 2002.

38. Brutsaert DL, Rademakers FE, Sys SU, Gillebert TC, Housman PR: Analysis of relaxation in the evaluation of ventricular function of the heart. Prog Cardiovasc Dis 28:143-163, 1985.

39. Sprung CL, Rackow EC, Fein IA, Jacob AI, Isikoff SK: The spectrum of pulmonary edema: Differentiation of cardiogenic, intermediate, and noncardiogenic forms of pulmonary edema. Am Rev Respir Dis 124:718-722, 1981.

40. Stang LB: Transalveolar capillary exchanges. In: Fishman AP, Hecht HH, eds. The Pulmonary Circulation and Interstitial Space. Chicago: University of Chicago Press; 1968:97-98.

41. Connors AF, Castele RJ, Farhat NZ, Tomashefski JF: Complications of right heart catheterisation. Chest 88:567-572, 1985.

Table 1. (A) Respiratory Score Classification and (B) Gait Score Classification and the Number and Percentage of Dogs of Each Score at Admission With Tick Paralysis

(A) Respiratory score Criteria No. Percentage

A Normal character and rate 9 22
(<30 breats/min)

B                                       Normal character and                   20                                      49
                                          increased rate
                                          (= or >30 breaths/min)

C Altered character with sigh/ 9 22
grunt (undefined rate)

D Cyanosis and severe dyspnea 3 7

Total 41 100

Adapted from National Tick Paralysis Forum, 1998.

(B) Gait score Criteria No. Percentage

1 Can walk; able to stand from 8 10
recumbency and ambulate

2                                        Cannot walk; requires aid to          9                                       22
                                          a standing position but can
                                          then maintain stance

3 Cannot stand; unable to 20 49
maintain a standing position

4 Cannot right; unable to maintain 4 10
sternal recumbency

Total 41 100

Adapted from Ilkiw JE: A Study of the Effects in the Dog of Ixodes holocyclus [PhD thesis]. University of Sydney, New South Wales; 1979.

Figure 1. Systolic blood pressure (mm Hg) measured at admission (n = 39), 24 hours (n = 26), discharge (n = 35), and 12 months (n = 11). The upper and lower ends of the box are the upper and lower quartiles between which the middle 50% of the data lies. The line dividing the box represents the median. The whiskers extending above and below the box show the highest and lowest values.

Figure 2. The artery:rib, vein:rib, and artery:vein ratios determined from thoracic radiographs of dogs at admission (n = 40), 24 hours (n = 26), and discharge (n = 35). A change in the artery:vein ratio was identified by repeated measures analysis of variance (P = 0.048) and the results of Tukey’s posttest are given. Bars represent the mean and their appendages represent the standard error of mean.

Figure 3. The peribronchial cuff:rib ratio measured from thoracic radiographs of dogs at admission (n = 18), 24 hours (n = 8), and discharge (n = 13). A change in the peribronchial cuff:rib ratio was identified by repeated measures analysis of variance (P = 0.01) and the results of Tukey’s posttest are given. Bars represent the mean and their appendages represent the standard error of mean.

Table 2. Mean (± standard error of mean) PCV and TP of Dogs at Admission, 24 Hours, and Discharge

Admission 24 hours Discharge

PCV (%) 51 ± 1A

(n = 40) 46 ± 0.9B

(n = 36) 44 ± 0.8C

(n = 35)

TP (g/ L) 71 ± 0.9A

(n = 38) 75 ± 1.3B

(n = 31) 74 ± 1.3B

(n = 35)

Values with different superscripts differ significantly.

Table 3. Mean (± standard error of mean) Values of Blood Parameters of Dogs at Admission and Discharge

Admission Discharge No.* P

PRA (ng/mL/hr) 1.14 ± 0.17 2.21 ± 0.44 26 (24) 0.03

PAC (ng/L) 61.6 ± 7.7 181.3 ± 34.5 26 (24) 0.003

PRA: PAC (hr-1) 21 ± 3 14 ± 2 24 0.049

Noradrenaline (nmol/L) 3.02 ± 0.77 1.29 ± 0.27 26 (24) 0.02

Adrenaline (nmol/L) 0.66 ± 0.14 0.48 ± 0.10 26 (24) 0.6

Cortisol (nmol/L) 184 ± 19 30 ± 1 24 (22) <0.0001

Lactate (mmol/L) 2.24 ± 0.20 2.01 ± 0.17 37 (34) 0.8

CTnT (µg/L) <0.01 <0.01 11 NT

LDH1 (arbitrary units) 5922 ± 798 4969 ± 965 20 0.3
*The number of values from which the mean is calculated. Parentheses enclose the numbers from which the mean is calculated at discharge when different from admission.

PRA, plasma renin activity; PAC, plasma aldosterone concentration; NT, not tested.

Table 4. Echocardiographic Measurements (mean ± standard error of mean) of Dogs at Admission, 24 Hours, and Discharge

Admission 24 hoursa Discharge

LVIDs (cm) 2.45 ± 0.09A 2.53 ± 0.11A 2.43 ± 0.12A

LVIDd (cm) 3.15 ± 0.12A 3.70 ± 0.16B 3.68 ± 0.14B

FS (%) 22 ± 1A 31 ± 1B 34 ± 1C

LVEsV (mL) 22.25 ± 2.25A 23.86 ± 2.66A 22.34 ± 3.01A

LVEdV (mL) 41.07 ± 3.87A 59.76 ± 5.58B 59.79 ± 5.41B

EF (%) 45 ± 3A 59 ± 3B 64 ± 2B

CO (mL/min) 2178 ± 244A 3668 ± 491B 3354 ± 294B

Heart rate (beats/min) 123 ± 4A 112 ± 5 94 ± 5B

Heart rate (beats/min)* (148 ± 6A) (120 ± 5) (112 ± 3B)

No. 39 26 35

Values with different superscripts differ significantly.

*Parentheses enclose values for heart rate assessed by auscultation before ACP.

†Assessed without ACP premedication.

Table 5. Signalment, Clinical Score and Edema Fluid to Serum Protein Ratio of Individual Dogs With Tick Paralysis

Breed (or primary cross) Age (y) Gait score Respiratory score Protein ratio*

Australian Cattle Dog 3 2 B 0.07

Rottweiler 12 4 C 0.20

Poodle 1 4 D 0.48

Labrador Unknown 4 D 0.14

Bull Mastiff 5 4 D 0.01†
*Edema fluid:serum protein ratio.

†Indicates treatment with 0.05 mg/kg ACP administered intravenously 20 to 40 minutes before sampling.