Abstract
OBJECTIVE
To determine whether administration of trimethoprim-sulfadiazine (TMS), detomidine (DET), or TMS plus DET would be associated with changes in ECG repolarization parameters in horses.
ANIMALS
9 healthy adult horses.
PROCEDURES
Each horse received 4 treatments in a blinded, randomized, crossover study design as follows: TMS, 16 to 24 mg/kg, IV; DET, 0.015 to 0.02 mg/kg, IV; TMS plus DET; and saline (0.9% NaCl) solution. Surface ECG traces were obtained over 24 hours, and repolarization parameters were measured at predefined time points after each treatment and compared with a 2-way ANOVA for repeated measures.
RESULTS
Heart rate–corrected QT intervals (QTc) were significantly increased after administration of DET (mean ± SD difference in QTc, 36.57 ± 23.07 milliseconds; increase of 7%) and TMS plus DET (44.96 ± 29.16 milliseconds; increase of 9%), compared with baseline (before treatment) values and values after administration of saline solution. Saline solution and TMS alone did not affect QTc.
CONCLUSIONS AND CLINICAL RELEVANCE
Administration of DET or TMS plus DET was associated with a significant and possibly clinically relevant prolongation of QTc, with prolongation of 7% to 9%, a range that is considered as a risk factor for the development of cardiac arrhythmias in people. Results were unexpected because DET is considered to be a safe sedative for horses.
Introduction
Case reports1,2,3 published in the mid-1980s described severe adverse effects such as dysrhythmias, hypotension, collapse, and death after concurrent IV administration of trimethoprim-sulfonamide compounds and DET. The mechanisms of these adverse effects were ostensibly never investigated in detail, but in several countries, recommendations were given to avoid IV administration of trimethoprim-sulfonamide compounds to sedated horses. Notwithstanding this limitation, DET is considered a safe sedative in horses.4,5 Trimethoprim-sulfonamide compounds, especially oral formulations, are widely prescribed antimicrobials for horses,6,7,8 and they are considered first− or second-line antimicrobials for the treatment of various bacterial infections.9,10 Trimethoprim-sulfonamide compounds are also available as IV formulations, which are useful when oral administration is not possible or rapid attainment of therapeutic drug concentrations is desirable.
Acquired long QT syndrome is well described in people11,12 and can be caused by drug interaction with cardiac potassium channels, especially the Kv11.1 channel (human ether-a-go-go-related gene [hERG] channel).11 Inhibition of cardiac potassium channels delays repolarization of the action potential, and this delay is reflected as prolonged QT interval on a surface ECG trace. Prolonged QT interval predisposes people to malignant arrhythmias and sudden death.11 Combinations of trimethoprim-sulfamethoxazole have been shown to prolong the QT interval in people13,14,15 and are not recommended for those with congenital long QT syndrome.16 Trimethoprim and TMS can block the cardiac Kv11.1 channels of horses in vitro.17 However, the effect of trimethoprim or TMS alone or in combination with DET on the QT interval in horses has not yet been reported. Assessing the effect of pharmaceuticals on the QT interval is required for human clinical safety studies, so pharmaceuticals with proarrhythmic potential are avoided.18,19 However, compounds that are known to inhibit the Kv11.1 channel in people are regularly used in veterinary medicine. Therefore, the aim of the study presented here was to investigate the effects of TMS and DET on the ECGs of healthy horses. Because TMS has been associated with aLQTS in people13,14,15 and has been shown to inhibit the Kv11.1 channels in vitro in horses,17 we hypothesized that TMS alone or in combination with DET would prolong the QT interval of healthy horses.
Materials and Methods
Animals
Nine horses (4 mares and 5 geldings) aged 3 to 18 years, of various breeds and types (Standardbred, 4; Friesian mix, 2; warmblood, 1; warmblood mix, 1; and Welsh Cob, 1), and body weight of 451 to 626 kg were available for the present study. All horses were considered healthy on the basis of results of physical examination, hematologic and serum biochemical analyses, and echocardiography. Horses were housed in box stalls with daily turnout and fed hay and concentrates according to resting demands. Horses were acclimated to the stall and barn for at least 1 month prior to the start of the study. Study design and procedures were approved by the Danish Animal and Experimentation Board (protocol No. 016-15-0201-00923) and the ethical committee at the study institution.
Study design
This randomized, blinded, crossover study was designed to investigate the effects of TMS with or without DET on the primary outcome of changes in the QTc at specified posttreatment time points, compared with baseline (pretreatment) and saline (0.9% NaCl) solution. To determine QTc, R-R intervals were measured. The secondary outcomes were drug effects on QRS duration, TpTe, TpTe/QT, and T-wave amplitude. The frequency of atrioventricular block was tabulated and serum potassium concentration was determined because these variables could also affect the QTc. Intra− and interobserver agreement for all ECG measurements was determined.
Experimental procedures
On experimental days, each horse was outfitted with a digital telemetric ECG devicea with surface electrodesb that were placed at anatomic locations as follows: dorsal aspect of the right scapula (red electrode [right arm negative]), approximately 5 cm from the red electrode on the dorsal aspect of the right scapula (black electrode [ground electrode]), over the intercostal space between the sixth and seventh ribs on the left side of the thorax approximately 30 cm ventral to the withers (yellow electrode [left arm positive]), and on the left side of the thorax at the level of the olecranon approximately 5 cm caudal to the xyphoid process (green electrode [left leg positive]).20 Each anatomic location was clipped free of hair, the skin over each was cleaned with alcohol, and electrodes were secured with sticking foam.c The ECG recording device was production calibrated, and ECGs were recorded with a sampling frequency of 500 Hz (x-axis) and a sampling resolution of 12 bit (y-axis). The ECGs were stored on a digital card for later analysis. For administration of drugs, an IV catheterd was placed into a jugular vein with an aseptic technique, alternating between jugular veins on the left side and right side (or vice versa) on each experiment day. Because hypokalemia can prolong the QT interval,11,12 a venous blood sample was obtained from the jugular catheter prior to each treatment and submitted to the Central Laboratory of the University of Copenhagen to determine serum potassium concentration with the ion-selective electrode method.e Each horse was kept in its stall and allowed to adapt to the ECG device for at least 15 minutes prior to treatment.
Each horse received the 4 treatments in a randomized, blinded, crossover design to minimize the number of horses included in this study. Treatment randomization was done in advance by drawing lots. Treatment 1 was TMSf at 16 to 24 mg/kg, IV, followed by a volume of saline solution,g IV, that was isovolumetric to the volume of DETh; treatment 2, saline solution, IV, that was isovolumetric to the volume of TMS, followed by DET at 0.015 to 0.02 mg/kg, IV; treatment 3, TMS followed by DET; and treatment 4, 2 doses of saline solution, IV, with one isovolumetric to TMS and the other isovolumetric to DET. Components of each treatment were administered 5 minutes apart.
Each treatment was administered at time 0 (between 10:00 am and 11:00 am), ECGs were recorded for 24 hours, and each horse was visually monitored continuously for the first 2 hours, then once every hour for 6 hours, and every 3 hours thereafter to assess its well-being and to check the position of the electrodes. After 24 hours, the ECG device, electrodes, and catheter were removed. A washout period of at least 3 days occurred between treatments, on the basis of drug withdrawal times and pharmacokinetic studies.21,22,23,24,25,26,27
ECG measurements
Each ECG trace to be analyzed had to be free of sudden transient tachycardia, second-degree atrioventricular block, and artifacts and had to correspond to a resting HR of < 50 bpm. For the designated time points of ECG analyses, ECG traces within 10 minutes before or after the designated time point could be analyzed to facilitate the selection of ECG traces that met the aforementioned criteria. Recordings of ECG traces were analyzed at these time points as follows: 10 minutes prior to treatment (baseline, −12 ± 5 minutes [mean ± SD]); then 5 (5 ± 1 minute), 15 (15 ± 1 minute), 30 (30 ± 3 minutes), 45 (45 ± 2 minutes), and 60 (58 ± 3 minutes) minutes; and every 30 minutes (89 ± 4 minutes, 119 ± 2 minutes, 148 ± 3 minutes, 178 ± 3 minutes, and 207 ± 4 minutes) after drug administration through 4 hours (238 ± 4 minutes), and then at 5 (298 ± 4 minutes), 6 (357 ± 5 minutes), 9 (538 ± 4 minutes), 12 (718 ± 4 minutes), and 22 hours (1,317 ± 20 minutes) after drug administration. For measurements, lead II ECG traces (corresponding to the signal registered between the right arm negative and left leg positive electrodes) were used. Traces were recorded at a paper speed of 200 mm/s and an amplitude of 40 mm/mV, and a 50-Hz filter was applied. Two investigators (DST and CSL) who were blinded to the treatments measured desired parameters with software for use with the ECG devicei and with an onscreen caliperj as described previously.28,2930,k At each time point, R-R intervals were measured for 10 consecutive beats immediately preceding the 5 complexes from which ECG measurements were taken to minimize the effects of QT hysteresis31 and then the mean was calculated. Heart rate was calculated by dividing 60,000 by the mean R-R interval (milliseconds). Durations of QRS, QT interval, QTc, and TpTe and peak T-wave amplitude were measured (Supplementary Figure S1, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.82.3.207). Duration of QRS was measured from the start of the Q wave to the point where the S wave intersected the isoelectric baseline. The QT interval was measured from the start of the Q wave to the end of the T wave where it intersected the isoelectric baseline. The threshold method, established as appropriate for horses,k was used to determine the end points of the S and T waves. Because QT interval depends on the HR, QT intervals were corrected for HR. A modified Framingham formula validated in horses by Petersen et al28,2930,k for a large range of HRs (including HRs at rest and during maximal exercise) was used to determine QTc with the following formula:
where ms is milliseconds and sloperest is the slope for HR < 60 bpm/min for Standardbreds28 and warmbloods.29 The formulas applied to calculate absolute (ΔQTc) and percentage point (%ΔQTc) differences in the QTc interval before and after treatment were as follows:
The T wave was characterized as monophasic or biphasic,28,29,30k and the amplitude of the peaks of the T wave was measured from the isoelectric line to the highest part of the T wave. The total T-wave amplitude was calculated by the following formula:
where peak 1 was the first deflection of the T wave (negative deflection) and peak 2 the second deflection (positive deflection). The time between the peak of the T wave and the end point of the T wave was calculated by the following formula:
where QTpeak was defined as the time between the start of the Q wave to the peak of the T wave or to the peak positive deflection of a biphasic T wave. The quotient of TpTe/QT was also calculated.
To determine whether the measurements were consistent, intraobserver variability (agreement) was determined for one of the investigators, an experienced veterinary cardiologist (DST), and interobserver variability was determined between the first and second investigators, the latter a veterinary student trained to obtain ECG measurements (CSL).
Statistical analysis
Graphical presentation, data analyses, and statistical analyses were performed with commercially available software.l,m,n Values of P ≤ 0.05 were considered significant. Normal distributions of the various ECG measurements were verified by visual assessment of histograms and normal probability plots of model residuals, and the Brown-Forsythe test was used to confirm equality of variance. Because the data were normally distributed and the assumption of equality of variance was fulfilled, parametric statistical methods were applied and data were reported as mean ± SD. Electrocardiographic measurements within (each time point vs baseline) and between (each treatment vs saline solution) treatment groups were compared with a 2-way ANOVA for repeated measures on the basis of a general linear model. In the model, the repeated level was horse and the independent explanatory factors were time point and drug. For correction of multiple comparisons, the Holm-Sidak post hoc test was used with baseline as the reference for time point and saline solution as the reference for drug. As a second step, a post hoc analysis was performed to detect differences between the treatments DET and TMS plus DET. To determine whether the order of the treatment day influenced the results (ie, first, second, third, or fourth treatment day, independent of treatment), a 2-way ANOVA based on a general linear model was used, including the independent explanatory factors drug and order of treatment day, for the data at baseline and 90 and 150 minutes. Because data were missing at the 5− and 15-minute time points, especially for DET and TMS plus DET, the statistical analyses were repeated excluding the data from these time points. The results of the repeated analysis did not differ from those of the first analysis (Supplementary Tables S1 and S2, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.82.3.207). Therefore, the results of the first analysis were reported. The changes in T-wave polarity, expressed as ordinal values, among treatment groups were compared with the Friedman test. Intra− and interobserver agreements were assessed through calculations of the coefficient of variation ([SD/mean] × 100%) and coefficient of repeatability (1.96 × √2 × SD of the difference between the 2 measurements) on the basis of the method described by Bland and Altman32,33 for randomly selected ECG traces from each time point (ECG traces for intraobserver agreement, n = 20; interobserver agreement, 16).
Results
None of the horses displayed signs of an adverse reaction during the observation period. The number of ECGs available for analysis for each of the 9 horses at each time point were summarized (Supplementary Table S3, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.82.3.207).
The general linear models indicated a significant interaction between the factors time point and drug within a treatment day for all of the ECG measurements except for TpTe and TpTe/QT. Additionally, ECG measurements returned to baseline during the observation period on each treatment day, which indicated that the possible treatment effects were transient, and the magnitude of the association seen for a treatment on an ECG measurement depended on the evaluated time point (Supplementary Table S3). Results of the analyses including order of treatment day also indicated that when a significant difference was detected, it was related to the factor drug and not the order of treatment day (Supplementary Table S4, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.82.3.207).
The R-R intervals significantly increased, which corresponded to significantly decreased HRs (Figure 1; Supplementary Table S5, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.82.3.207), from baseline at 5, 15, and 30 minutes after administration of DET and at 5 and 30 minutes after administration of TMS plus DET. Also, R-R intervals for DET were significantly increased from those for saline solution at 5, 15, 30, 45, and 60 minutes and for TMS plus DET at 5, 15, 30, and 45 minutes. In contrast, R-R intervals were not significantly different after administration of saline solution or TMS. Within the first 30 minutes after administration of DET or TMS plus DET, frequent second-degree atrioventricular block was observed. At 240, 300, 360, and 540 minutes after administration of TMS plus DET, R-R intervals were significantly decreased (HR significantly increased), compared with baseline.
Mean ± SD for R-R interval (A), QRS duration (B), QTc (C), and difference in QTc (ΔQTc; D) for 9 healthy adult horses obtained before (baseline [BAS]) and at various time points after 5, 15, 30, 45, 60, 90, 120, 150, 180, 210, 240, 300, 360, 540, 720, and 1,320 minutes) administration 1 of 4 treatments as follows: saline (0.9% NaCl) solution (circles), TMS (squares), DET (triangles), and TMS plus DET (diamonds). In panel C, the dashed line represents the reference range for QTc of horses.29 ΔQTc = (QTctime point – QTcBAS). *Indicates a significant (P ≤ 0.05) difference between treatment and saline solution for the designated time point. Each bracket indicates a significant (P ≤ 0.05) difference between baseline and the designated time point within each treatment.
Citation: American Journal of Veterinary Research 82, 3; 10.2460/ajvr.82.3.207
Compared with baseline, QT intervals were decreased (shorter) for all treatments after 360 minutes and were significantly decreased for TMS, DET, and TMS plus DET between 360 and 720 minutes after drug administration (data not shown). Additionally, QT intervals were significantly prolonged for DET and TMS plus DET from 5 to 180 minutes after drug administration (data not shown). Because the QT interval depended on the HR and HR was affected by DET and TMS plus DET, QT intervals were corrected for HR (QTc) and only QTc values were used for further analyses. Compared with baseline and saline solution, QTc was significantly prolonged between 45 and 90 minutes after administration of DET and between 30 and 210 minutes after TMS plus DET (Figure 1; Supplementary Table S6, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.82.3.207). The largest prolongation of the QTc occurred 45 minutes after DET (ΔQTc: mean ± SD, 36.57 ± 23.07 milliseconds; 95% CI, 15.2 to 57.9 milliseconds) or TMS plus DET (44.96 ± 29.16 milliseconds; 20.6 to 69.3 milliseconds) administration, which corresponded to prolongation of 7% and 9%, respectively (Supplementary Tables S7 and S8, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.82.3.207). Shortening of the QTc was found 360 to 720 minutes after drug administration, but compared with the results from uncorrected QT intervals, the effect was only significant for TMS. Post hoc analyses revealed no significant differences in all measured ECG parameters between DET or TMS plus DET.
T waves were consistently biphasic (n = 6 horses) or changed from biphasic to a single positive (2) or negative (1) deflection. Treatment did not significantly (P = 0.33) affect T-wave morphology. Peak 1 of the T wave was significantly more negative with DET versus baseline at 30 and 45 minutes and versus saline solution at 5, 15, 30, 45, 60, 150, 180, and 210 minutes (Figure 2; Supplementary Table S9, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.82.3.207). Similarly, peak 1 was significantly more negative with TMS plus DET versus baseline at 30 minutes and versus saline solution at 5, 15, 30, 45, 150, and 180 minutes. Peak 2 of the T wave was significantly less positive at 45 and 60 minutes with DET and 30 and 45 minutes with TMS plus DET, compared with saline solution (Supplementary Table S10, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.82.3.207). Values for total T wave were summarized (Supplementary Table S11, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.82.3.207).
Mean ± SD total T-wave amplitude (A), first (negative [neg]) T-wave peak (B), and second (positive [pos]) T-wave peak (C) for the horses of Figure 1. Total T-wave amplitude is the sum of the amplitudes of the 2 peaks (total T-wave amplitude = [–peak 1] + [peak 2], where peak 1 is the first [negative] deflection and peak 2 the second [positive] deflection.) See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 82, 3; 10.2460/ajvr.82.3.207
No significant differences were noted for TpTe within and between treatments (Figure 3; Supplementary Table S12, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.82.3.207). The quotient of TpTe/QT was significantly less at 30 and 45 minutes for DET and 30, 45, and 60 minutes for TMS plus DET, compared with saline solution (Supplementary Table S13, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.82.3.207).
Mean ± SD TpTe (A) and TpTe/QT (B) for the horses of Figure 1. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 82, 3; 10.2460/ajvr.82.3.207
The effects of TMS, DET, or TMS plus DET on QRS duration were inconsistent (Supplementary Table S14, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.82.3.207). Mean (± SD) serum potassium concentration was 3.6 (± 0.8 mmol/L), and all values were within the reference range.
Intra− and interobserver agreements for the ECG measurements were summarized (Table 1). Coefficients of variation were highest for QRS duration, TpTe, and peaks 1 and 2 of the T wave. The coefficient of repeatability for the QT interval and QTc varied between approximately 21 and 31 milliseconds.
Intra− and interobserver agreement for the measurement of select parameters obtained from randomly selected lead II ECG traces (intraobserver agreement, n = 20 ECG traces; interobserver agreement, 16) recorded from 9 healthy adult horses before and at various time points after IV administration of 1 of 4 treatments (saline solution, TMS, DET, or TMS plus DET).
| CV (%) | SD of difference between measurements | Coefficient of repeatability | ||||
|---|---|---|---|---|---|---|
| ECG parameter | Intraobserver | Interobserver | Intraobserver | Interobserver | Intraobserver | Interobserver |
| R-R interval (ms) | 1.13 | 1.96 | 30.74 | 46.57 | 85.14 | 129.00 |
| HR (bpm) | 1.13 | 1.96 | 0.76 | 1.40 | 2.11 | 3.88 |
| QRS duration (ms) | 3.84 | 5.19 | 3.38 | 7.74 | 9.35 | 21.43 |
| QT interval (ms) | 0.93 | 2.47 | 7.49 | 11.10 | 20.75 | 30.77 |
| QTc (ms) | 1.08 | 2.41 | 7.70 | 8.68 | 21.33 | 24.04 |
| TpTe (ms) | 10.03 | 11.12 | 7.96 | 5.77 | 22.06 | 15.99 |
| T wave peak 1 (mV) | 4.55 | 17.86 | 0.02 | 0.07 | 0.06 | 0.21 |
| T wave peak 2 (mV) | 3.75 | 12.05 | 0.02 | 0.06 | 0.04 | 0.16 |
Discussion
Results indicated that cardiac repolarization based on QT interval and T-wave morphology determined from lead II ECG traces may be affected in horses after administration of DET or TMS plus DET; however, QRS duration was minimally and inconsistently affected by treatment. The results of the present study contrasted with the few reports of adverse effects when DET was used alone. Further studies are needed to determine what other predisposing factors are necessary to precipitate severe adverse effects reported1,2,3 after administration of TMS in combination with DET.
The QT intervals were corrected for HR (QTc), and results indicated that prolongation, a possible drug effect, was not caused by a decrease in HR. Compared with baseline, the largest changes in the QTc were noted with DET and TMS plus DET 45 minutes after their administration, which represented a 7% to 9% increase (prolongation). In people, an increase of > 20 milliseconds for QTc, approximately a 4% increase from the normal QTc, is considered a risk factor for torsade de pointes arrhythmias.15 Therefore, an increase of 7% to 9% may have clinical relevance in horses. Yet, even if a prolonged QTc in people is the hallmark of aLQTS and a risk factor for development of malignant arrhythmias and sudden cardiac death,11,12,18,19 these are rare and likely also depend on the genetic subtype of Kv11.1 that is present.34 Therefore, further studies should likewise assess such genetic susceptibility in horses.
Arrhythmias are rarely reported among the cardiovascular adverse effects of DET in horses,4,5 and α2-adrenoceptor agonists may be protective against epinephrine-induced arrhythmias in dogs anesthetized with halothane.35 The horses in the present study were at rest in an environment to which they were acclimated for at least 1 month, so the extent to which DET interferes with the physiologic adaptation of the duration of the action potential and the repolarization reserve36 in a stressful situation, as previously reported,1,2,3 needs to be further investigated.
Acquired long QT syndrome in people is often caused by inhibition of the voltage-gated potassium channel Kv11.1. Trimethoprim-sulfadiazine has been previously17,37 demonstrated to block equine Kv11.1 expressed in the oocytes of Xenopus laevis and in mammalian cells. In both studies,17,37 clinically relevant concentrations of TMS were reached after its IV administration decreased the Kv11.1 current by 30% to 50%, which indicated that TMS could cause aLQTS in horses. However, no changes in the assessed parameters of cardiac repolarization were identified in the present in vivo study after IV administration of a standard dosage of TMS. One possible explanation for the difference among studies is that the plasma concentration of TMS attained after IV administration of a standard dose to healthy horses may not reach a concentration that causes clinically relevant inhibition of Kv11.1. In one of the previous studies,17 the IC50 was 3.7mM for TMS on equine Kv11.1 expressed in Xenopus laevis oocytes, and in the other study,37 the IC50 was 100μM on equine Kv11.1 expressed in mammalian cells. The differences between the IC50 values obtained in those 2 heterologous systems are likely due to binding of TMS to the oocyte yolk.38 The IC50 for TMS on heterologously expressed equine Kv11.1 channels were higher than the plasma concentrations of 12μM to 34μM attained after IV administration of a single dose.17,23,24,25 However, a large safety margin between the IC50 and free plasma concentration has been advocated39; therefore, a 30− to 40-fold IC50 versus free plasma concentration ratio has to be demonstrated before a drug can be judged as safe.39,40 For a drug like TMS that is protein bound between 35% and 50%,3,24 the estimated free plasma concentration of TMS reaches values approximately a thirtieth to a fortieth of the IC50,17 suggesting that the difference between free plasma concentration after a standard dosage and IC50 would not be large enough and could affect Kv11.1 function. However, the sensitivity of Kv11.1 to TMS may vary among horses, such that the effect of TMS may also vary. Therefore, in the present study, higher doses of TMS may have been necessary to induce an effect on the QT interval. Additionally, several potassium channels are involved in cardiac repolarization, resulting in a repolarization reserve.36 These redundant potassium channels and others may compensate for a partially blocked Kv11.1 current.36 Thus, differences in repolarization reserve or genetic variants may make some individuals more or less sensitive to the drug effect.
The results of the present study indicated that prolongation of the QT interval and QTc after administration of DET or TMS plus DET was caused by DET. The timing of an identifiable prolonged QT interval was concordant with previous reports.21,22 Administration of a similar dose of DET (30 μg/kg) resulted in a decrease in the HR for approximately 90 minutes, changes in mucous membrane color for approximately 120 minutes, and increased capillary refill time for approximately 180 minutes.21,22 The initial cardiovascular effect of DET has been attributed mainly to the activation of α2B-adrenoceptors located on blood vessels, causing transient vasoconstriction followed by a rise in blood pressure and reflex bradycardia.41,42 Activation of α2A-adrenoceptors is likely responsible for a later decrease in blood pressure.41,42 Cross-activation of α1-adrenoceptors may lead to vasoconstriction but is likely less important in the cardiovascular response to DET because the affinity of DET to α1-adrenoceptors is comparatively low (affinity ratio of α1:α2, 1:260).43 Plus, the activation of α1-adrenoceptors is more likely indicative of an adverse reaction, such as excitement, with high doses.42
The analgesic and sedative effects of DET have been documented as a reduced response to nociceptive stimuli for approximately 90 minutes and a low head position for 240 minutes following its administration.21,22,44 The analgesic effects are mainly related to activation of α2A-, α2B-, and α2C-adrenoceptors in the spinal cord, whereas the sedative effects of DET are mainly mediated by activation of α2-adrenoceptors located in the supraspinal portion of the CNS.41,42 The effects of α2A-adrenoceptor agonists are closely related to the negative feedback loop for autoregulation of the sympathetic nervous system, resulting in reduced release of norepinephrine and reduced central sympathetic outflow.41,42 Even if the HR is primarily regulated by the parasympathetic nervous system at rest,45 low-grade sympathetic simulation is present and, moreover, blocking sympathetic stimulation results in a small reduction in resting HR.46,47,48 Besides Kv11.1, the voltage-gated potassium channel Kv7.1 is important for cardiac repolarization in horses.49,50,51 The activity of Kv7.1 depends on sympathetic stimulation of the heart and activation of β1-adrenoceptors, and increased Kv7.1 currents result in concomitant shortening of the durations of the action potential and QT interval during sympathetic stimulation.52,53 This is important for adaptation of the action potential duration to fast heart rates, as during exercise or stress. Cross-activation of β1-adrenoceptors is unlikely to occur with α2-adrenoceptor agonists,43 but the observed effect of DET could possibly be explained by an α2-adrenoceptor agonist–mediated reduction in central output of the sympathetic drive, resulting in reduced intrinsic activity of Kv7.1 and, therefore, prolongation of the durations of action potential and QT interval. This supposition agrees with an observed small increase in the QT interval at low HRs in people treated with β-adrenoceptor antagonists.54
Shortening of the QT interval and QTc between 6 and 12 hours after drug administration was observed for all treatments, although the difference was not significant after saline solution administration. On the basis of the timing of drug administration, this finding was similar with the diurnal variation of the QT interval reported in a previous study,30 in which the shortest QT intervals occurred during the evening and early overnight hours. The inconsistent changes in the QRS duration suggested that TMS, DET, or TMS plus DET do not affect cardiac conduction velocity.
To further assess the association between DET or TMS and repolarization, TpTe, TpTe/QT, and T-wave morphology were analyzed. In human medicine, these parameters have been frequently evaluated because of their potential to help predict the development of torsade de pointes arrythmias associated with long QT syndrome.55,56,57,58 The TpTe, which represents total dispersion of ventricular repolarization, is related to the inhomogeneity of ventricular repolarization that is partially caused by the distribution of ion channels in various parts of the heart, leading to the heart's electrical heterogeneity.59 The advantage of TpTe versus QT interval is that TpTe is likely independent of HR.56 Heart rate independence for TpTe has been shown for mares28 and for monophasic T waves for geldings,29 but diurnal variation has also been reported30; therefore, TpTe is not a consistent measure in horses. Furthermore, the frequent occurrence of biphasic T waves in horses makes the TpTe challenging to interpret and comparison among species difficult. In the present study, the drugs had no effect on the TpTe and the measured changes for TpTe/QT over time were small. These findings may have indicated that the drugs similarly affect all areas of the ventricle55,60 and do not influence electrical heterogeneity despite increased QTc. This may have also indicated that aLQTS is less problematic in horses, compared with people, because dispersion of ventricular repolarization is less. The resulting more homogenous repolarization in the equine heart might protect against arrythmias. However, more studies are needed before this can be concluded.
The intra− and interobserver CVs for ECG measurements were low and corresponded to those previously reported28,29,30,k for horses. The highest CVs were noted for QRS duration, TpTe, and peak T-wave amplitude. Measurements of QRS duration and peak amplitude of the T wave rely on a stable baseline between the Q wave and the end point of the T wave. However, this baseline may drift, resulting in measurement differences between observers. Yet, CVs for all ECG measurements in the present study were < 25%, which is acceptable for other techniques61 and similar to CVs reported62,63,64 for echocardiographic measurements of horses. The SD of intra− and interobserver differences was similar to that reported65,66,67 in people. Calculated coefficients of repeatability for ECG measurements were similar to the 95% limit of agreement values identified with Bland-Altman analysis, which indicated that the lack of agreement between observers might explain the difference.33 This finding highlighted the need for proper training in taking ECG measurements to increase data quality and possibly increase the likelihood of detecting prolongation of the QT interval.68
One of the main limitations of the present study was the lack of thoroughly validated reference values for QT and QTc intervals in horses28,29,30 and of knowledge about action potential repolarization and T-wave morphology in horses. The published formula for QTc is based on a small number of horses and should be validated with a larger sample size.28,29,30 The HR value for use with the formula is limited to HRs > 25 bpm, and this restriction may limit the formula's use, despite that in the present study only 1 horse at 1 time point had an HR < 25 bpm. Furthermore, ECG parameters were measured at predetermined time points rather than continuously. Maximum response for each horse may have occurred between these time points, resulting in an underestimation of the measured changes. Additionally, the horses were of various ages and breeds, and correction formulas were used to account for any breed differences, but their use may not have eliminated all individual differences. Individual sensitivity to drug effects and the dose range used could have also affected the results. Because of financial and ethical concerns, including more horses and > 8 treatment arms was not possible. A carryover effect after IV administration of a single dose of TMS and DET in healthy horses after a washout period of at least 3 days, on the basis of withdrawal time and pharmacokinetic studies,21,22,23,24,25,26,27 was unlikely to have affected our results. However, the small number of horses resulted in reduced power for some analyses. Therefore, a type 2 error cannot be excluded and larger studies will be necessary to confirm our results.
The main findings of this preliminary study were that DET alone or combined with TMS prolonged QTc, with prolongation outside of the reference range for horses,28,29 and TMS alone had no effect on QTc. The QTc prolongation represented a 7% to 9% increase versus predrug QTc. Furthermore, no drug affected QRS duration. These results contrasted with the broad and safe clinical use of DET and the few reported adverse effects of this drug. However, reports of death after administration of TMS plus DET have been published,1,2,3 and sudden cardiac death in relation to medical treatment is likely underreported in veterinary medicine, compared with human medicine.69,70 Taken together with the results of the present study, aLQTS should be considered as a differential diagnosis in cases of cardiovascular incidents, particularly in relation to drug administration. In people, aLQTS is associated with ventricular arrhythmias, such as torsade de pointes, and sudden cardiac death and has led to withdrawal of several drugs from the markets of many countries40 and to the development of extensive cardiac safety programs.18,19 Yet, the risk of death related to long QT syndrome is low in people.34 Conducting further studies on cardiac repolarization in horses is important to determine whether prolonged QT interval secondary to drug administration is associated with an increased risk of developing malignant arrhythmias in horses as in people before any definitive conclusion about the proarrhythmic capability of TMS (or DET or TMS plus DET) can be made.
Acknowledgments
Funded by the European Union's Horizon 2020 research and innovation program (Marie Sklodowska Curie grant agreement No. 656566) and the foundation Foreningen KUSTOS af 1881.
The authors declare that there were no conflicts of interest.
Presented as a poster at the 8th Nordic Equine Veterinary Conference, Blomsterdalen, Norway 2018.
The authors thank Dr. Adrian Paul Harrison for reviewing the manuscript.
Abbreviations
| aLQTS | Acquired long QT syndrome |
| bpm | Beats per minute |
| CV | Coefficient of variation |
| DET | Detomidine |
| HR | Heart rate |
| IC50 | Half maximal inhibitory concentration |
| QTc | QT interval corrected for HR |
| TMS | Trimethoprim-sulfadiazine |
| TpTe | Time between peak and end of the T wave |
Footnotes
TELEVET 100, Engel Engineering Service GmbH, Offenbach am Main, Germany.
KRUUSE AS, Maarslev, Denmark.
Snögg AS, Mosby, Norway.
Milacath, E-VET A/S, Haderslev, Denmark.
Advia 1800 Chemistry System, Siemens Healthcare A/S, Ballerup, Denmark.
Noradine VET containing a trimethoprim/sulfadiazine ratio of 40/200 mg, ScanVet Animal Health A/S, Fredensborg, Denmark.
B Braun Melsungen AG. Melsungen, Germany.
Domosedan VET, Orion Pharma Animal Health, Copenhagen, Denmark.
TELEVET 100 software, version 6.0 Engel Engineering Services GmbH, Offenbach am Main, Germany.
Cardio Calipers 3,3, ICONICo Inc, New York, NY.
Pedersen PJ. Building a platform for diagnosing long QT syndrome in horses. PhD thesis, Department of Veterinary and Animal Sciences, University of Copenhagen, Frederiksberg C, Denmark, 2014.
GraphPad Prism, version 7.00, GraphPad Software Inc, La Jolla, Calif.
SigmaStat, version 3.5, Systat Software GmbH, Erkrath, Germany.
Microsoft Excel, Office 2010, Redmond, Wash.
References
- 1. ↑
Dick IG, White SK. Possible potentiated sulphonamide-associated fatality in an anaesthetised horse. Vet Rec 1987;121:288.
- 2. ↑
Taylor PM, Rest RJ, Duckham TN, et al. Possible potentiated sulphonamide and detomidine interactions. Vet Rec 1988;122:143.
- 3. ↑
Van Duijkeren E, Vulto AG, Van Miert AS. Trimethoprim/sulfonamide combinations in the horse: a review. J Vet Pharmacol Ther 1994;17:64–73.
- 4. ↑
Hubbell JAE. Horses. In: Tranquilli WJ, Thurmon JC, Grimm KA, eds. Lumb and Jones’ veterinary anesthesia and analgesia. 4th ed. Ames, Iowa: Blackwell Publishing, 2007;717–729.
- 5. ↑
Daunt DA, Steffey EP. Alpha-2 adrenergic agonists as analgesics in horses. Vet Clin North Am Equine Pract 2002;18:39–46.
- 6. ↑
Haggett EF, Wilson WD. Overview of the use of antimicrobials for the treatment of bacterial infections in horses. Equine Vet Educ 2008;20:433–448.
- 7. ↑
Hughes LA, Pinchbeck G, Callaby R, et al. Antimicrobial prescribing practice in UK equine veterinary practice. Equine Vet J 2013;45:141–147.
- 8. ↑
Regula G, Torriani K, Gassner B, et al. Prescription patterns of antimicrobials in veterinary practices in Switzerland. J Antimicrob Chemother 2009;63:805–811.
- 9. ↑
Morley PS, Apley MD, Besser TE, et al. Antimicrobial drug use in veterinary medicine. J Vet Intern Med 2005;19:617–629.
- 10. ↑
Weese JS, Baptiste KE, Baverud V, et al. Guidelines for antimicrobial use in horses. In: Guardabassi L, Jensen LB, Kruse H, eds. Guide to antimicrobial use in animals. Oxford: Blackwell Publishing Ltd, 2008;161–182.
- 11. ↑
Kallergis EM, Goudis CA, Simantirakis EN, et al. Mechanisms, risk factors, and management of acquired long QT syndrome: a comprehensive review. Sci World J 2012;2012:212178.
- 12. ↑
Roden DM. Acquired long QT syndromes and the risk of pro-arrhythmia. J Cardiovasc Electrophysiol 2000;11:938–940.
- 13. ↑
Lopez JA, Harold JG, Rosenthal MC, et al. QT prolongation and torsades de pointes after administration of trimethoprim-sulfamethoxazole. Am J Cardiol 1987;59:376–377.
- 14. ↑
Wiener I, Rubin DA, Martinez E, et al. QT prolongation and paroxysmal ventricular tachycardia occurring during fever following trimethoprim-sulfamethoxazole administration. Mt Sinai J Med 1981;48:53–55.
- 15. ↑
Shah RR. The significance of QT interval in drug development. Br J Clin Pharmacol 2002;54:188–202.
- 16. ↑
Woosley RL, Heise CW, Romero KA. QT drugs list. Oro Valley, Ariz: AZCERT Inc, 2017. Available at: www.crediblemeds.org/. Accessed Nov 10, 2017.
- 17. ↑
Trachsel DS, Tejada MA, Groesfjeld Christensen V, et al. Effects of trimethoprim-sulfadiazine and detomidine on the function of equine Kv 11.1 channels in a two-electrode voltage-clamp (TEVC) oocyte model. J Vet Pharmacol Ther 2018;41:536–545.
- 18. ↑
Expert working group (efficacy) of the international conference on harmonisation of technical requirements for registration of pharmaceuticals for human use (ICH). Guidance for industry: E14 clinical evaluation of QT/QTc interval prolon gation and proarrhythmic potential for non-antiarrhythmic drugs. Rockville, Md: FDA, 2005. Available at: www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm073153.pdf. Accessed Jun 5, 2018.
- 19. ↑
Committee for medicinal products for human use. ICH Topic E 14: the clinical evaluation of QT/QTc interval prolongation and proarrhythmic potential for non-antiarrhythmic drugs. Amsterdam: European Medicines Agency, 2005. Available at: www.ema.europa.eu/ema/index.jsp?curl=pages/regulation/general/general_content_001081.jsp&mid=WC0b01ac0580034cef. Accessed Jun 5, 2018.
- 20. ↑
Fleth⊘j M, Kanters JK, Haugaard MM, et al. Changes in heart rate, arrhythmia frequency, and cardiac biomarker values in horses during recovery after a long-distance endurance ride. J Am Vet Med Assoc 2016;248:1034–1042.
- 21. ↑
Grimsrud KN, Ait-Oudhia S, Durbin-Johnson BP, et al. Pharmacokinetic and pharmacodynamic analysis comparing diverse effects of detomidine, medetomidine, and dexmedetomidine in the horse: a population analysis. J Vet Pharmacol Ther 2015;38:24–34.
- 22. ↑
Mama KR, Grimsrud K, Snell T, et al. Plasma concentrations, behavioural and physiological effects following intravenous and intramuscular detomidine in horses. Equine Vet J 2009;41:772–777.
- 24. ↑
Gustafsson A, Baverud V, Franklin A, et al. Repeated administration of trimethoprim/sulfadiazine in the horse–pharmacokinetics, plasma protein binding and influence on the intestinal microflora. J Vet Pharmacol Ther 1999;22:20–26.
- 25. ↑
van Duijkeren E, Vulto AG, Sloet van Oldruitenborghoosterbaan MM, et al. A comparative study of the pharmacokinetics of intravenous and oral trimethoprim/sulfadiazine formulations in the horse. J Vet Pharmacol Ther 1994;17:440–446.
- 26. ↑
EHSLC detection times July 2015. London: The European Horserace Scientific Liaison Committee (EHSLC), 2015. Available at: www.ehslc.com/detection-times. Accessed Jun 5, 2018.
- 27. ↑
Controlled therapeutic substance list. Lexington, Ky: Racing Medication and Testing Consortium, 2016. Available at: www.rmtcnet.com/national-uniform-medication-program/controlled-therapeutic-substances/#. Accessed Jun 5, 2018.
- 28. ↑
Pedersen PJ, Kanters JK, Buhl R, et al. Normal electrocardiographic QT interval in race-fit Standardbred horses at rest and its rate dependence during exercise. J Vet Cardiol 2013;15:23–31.
- 29. ↑
Pedersen PJ, Karlsson M, Fleth⊘j M, et al. Differences in the electrocardiographic QT interval of various breeds of athletic horses during rest and exercise. J Vet Cardiol 2016;18:255–264.
- 30. ↑
Pedersen PJ, Moeller SB, Fleth⊘j M, et al. Diurnal modulation and sources of variation affecting ventricular repolarization in Warmblood horses. J Vet Cardiol 2014;16:265–276.
- 31. ↑
Malik M, Hnatkova K, Novotny T, et al. Subject-specific profiles of QT/RR hysteresis. Am J Physiol Heart Circ Physiol 2008;295:H2356–H2363.
- 32. ↑
Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307–310.
- 33. ↑
Bland JM, Altman DG. Measuring agreement in method comparison studies. Stat Methods Med Res 1999;8:135–160.
- 34. ↑
Priori SG, Schwartz PJ, Napolitano C, et al. Risk stratification in the long-QT syndrome. N Engl J Med 2003;348:1866–1874.
- 35. ↑
Hayashi Y, Sumikawa K, Maze M, et al. Dexmedetomidine prevents epinephrine-induced arrhythmias through stimulation of central alpha 2 adrenoceptors in halothane-anesthetized dogs. Anesthesiology 1991;75:113–117.
- 36. ↑
Roden DM. Long QT syndrome: reduced repolarization reserve and the genetic link. J Intern Med 2006;259:59–69.
- 37. ↑
Calloe K, Rognant S, Friis S, et al. Compounds commonly used in equine medicine inhibits the voltage-gated potassium channel Kv11.1. Res Vet Sci 2019;123:239–246.
- 38. ↑
Witchel HJ, Milnes JT, Mitcheson JS, et al. Troubleshooting problems with in vitro screening of drugs for QT interval prolongation using HERG K+ channels expressed in mammalian cell lines and Xenopus oocytes. J Pharmacol Toxicol Methods 2002;48:65–80.
- 39. ↑
Redfern WS, Carlsson L, Davis AS, et al. Relationships between preclinical cardiac electrophysiology, clinical QT interval prolongation and torsade de pointes for a broad range of drugs: evidence for a provisional safety margin in drug development. Cardiovasc Res 2003;58:32–45.
- 40. ↑
Gintant G. An evaluation of hERG current assay performance: Translating preclinical safety studies to clinical QT prolongation. Pharmacol Ther 2011;129:109–119.
- 41. ↑
Philipp M, Brede M, Hein L. Physiological significance of alpha(2)-adrenergic receptor subtype diversity: one receptor is not enough. Am J Physiol Regul Integr Comp Physiol 2002;283:R287–R295.
- 42. ↑
Lemke KA. Anticholinergics and Sedatives In: Tranquilli WJ, Thurmon JC, Grimm KA, eds. Lumb and Jones’ veterinary anesthesia and analgesia. 4th ed. Ames, Iowa: Blackwell Publishing, 2007;203–240.
- 43. ↑
Virtanen R, Savola JM, Saano V, et al. Characterization of the selectivity, specificity and potency of medetomidine as an alpha 2-adrenoceptor agonist. Eur J Pharmacol 1988;150:9–14.
- 44. ↑
Jöchle W, Hamm D. Sedation and analgesia with Domosedan (detomidine hydrochloride) in horses: dose response studies on efficacy and its duration. Acta Vet Scand Suppl 1986;82:69–84.
- 45. ↑
Gordan R, Gwathmey JK, Xie LH. Autonomic and endocrine control of cardiovascular function. World J Cardiol 2015;7:204–214.
- 46. ↑
Robinson BF, Epstein SE, Beiser GD, et al. Control of heart rate by the autonomic nervous system. Studies in man on the interrelation between baroreceptor mechanisms and exercise. Circ Res 1966;19:400–411.
- 47. ↑
Chamberlain DA, Howard J. The haemodynamic effects of beta-sympathetic blockade. Br Heart J 1964;26:213–217.
- 48. ↑
Glick G, Braunwald E. Relative roles of sympathetic and parasympathetic nervous systems in reflex control of heart rate. Circ Res 1965;16:363–375.
- 49. ↑
Pedersen PJ, Thomsen KB, Flak JB, et al. Molecular cloning and functional expression of the K+ channel KV7.1 and the regulatory subunit KCNE1 from equine myocardium. Res Vet Sci 2017;113:79–86.
- 50. ↑
Pedersen PJ, Thomsen KB, Olander ER, et al. Molecular cloning and functional expression of the equine K+ channel KV11.1 (Ether à go-go-related/KCNH2 gene) and the regulatory subunit KCNE2 from equine myocardium. PLoS One 2015;10:e0138320.
- 51. ↑
Finley MR, Li Y, Hua F, et al. Expression and coassociation of ERG1, KCNQ1, and KCNE1 potassium channel proteins in horse heart. Am J Physiol Heart Circ Physiol 2002;283:H126–H138.
- 52. ↑
Terrenoire C, Clancy CE, Cormier JW, et al. Autonomic control of cardiac action potentials: role of potassium channel kinetics in response to sympathetic stimulation. Circ Res 2005;96:e25–e34.
- 53. ↑
Wu J, Ding WG, Horie M. Molecular pathogenesis of long QT syndrome type 1. J Arrhythm 2016;32:381–388.
- 54. ↑
Bennett MT, Gula LJ, Klein GJ, et al. Effect of beta-blockers on QT dynamics in the long QT syndrome: measuring the benefit. Europace 2014;16:1847–1851.
- 55. ↑
Yamaguchi M, Shimizu M, Ino H, et al. T wave peak-to-end interval and QT dispersion in acquired long QT syndrome: a new index for arrhythmogenicity. Clin Sci (Lond) 2003;105:671–676.
- 56. ↑
Kanters JK, Haarmark C, Vedel-Larsen E, et al. T(peak)T(end) interval in long QT syndrome. J Electrocardiol 2008;41:603–608.
- 57. ↑
Kors JA, Ritsema van Eck HJ, van Herpen G. The meaning of the Tp-Te interval and its diagnostic value. J Electrocardiol 2008;41:575–580.
- 58. ↑
Castro-Torres Y, Carmona-Puerta R, Katholi RE. Ventricular repolarization markers for predicting malignant arrhythmias in clinical practice. World J Clin Cases 2015;3:705–720.
- 59. ↑
Xia Y, Liang Y, Kongstad O, et al. Tpeak-Tend interval as an index of global dispersion of ventricular repolarization: evaluations using monophasic action potential mapping of the epi− and endocardium in swine. J Interv Card Electrophysiol 2005;14:79–87.
- 60. ↑
Roche BM, Kijtawornrat A, Hamlin RL, et al. Relationship between prolongation of QTc and prolongation of the peak of T (Tp) to the end of T (Te). J Pharmacol Toxicol Methods 2005;52:178–181.
- 61. ↑
Valentin MA, Ma S, Zhao A, et al. Validation of immunoassay for protein biomarkers: bioanalytical study plan implementation to support pre-clinical and clinical studies. J Pharm Biomed Anal 2011;55:869–877.
- 62. ↑
Buhl R, Ersboll AK, Eriksen L, et al. Sources and magnitude of variation of echocardiographic measurements in normal standardbred horses. Vet Radiol Ultrasound 2004;45:505–512.
- 63. ↑
Schwarzwald CC, Schober KE, Bonagura JD. Methods and reliability of echocardiographic assessment of left atrial size and mechanical function in horses. Am J Vet Res 2007;68:735–747.
- 64. ↑
Schwarzwald CC, Schober KE, Bonagura JD. Methods and reliability of tissue Doppler imaging for assessment of left ventricular radial wall motion in horses. J Vet Intern Med 2009;23:643–652.
- 65. ↑
Natekar M, Mahajan V, Satra A, et al. Use of ANOVA to estimate inter− and intra-reader variability for a group of readers in thorough QT/QTc studies. Clin Pharmacol Ther 2008;83:489–491.
- 66. ↑
Panicker GK, Karnad DR, Natekar M, et al. Intra− and interreader variability in QT interval measurement by tangent and threshold methods in a central electrocardiogram laboratory. J Electrocardiol 2009;42:348–352.
- 67. ↑
Salvi V, Karnad DR, Kerkar V, et al. Comparison of two methods of estimating reader variability in QT interval measurements in thorough QT/QTc studies. Ann Noninvasive Electrocardiol 2014;19:182–189.
- 68. ↑
Viskin S, Rosovski U, Sands AJ, et al. Inaccurate electrocardiographic interpretation of long QT: the majority of physicians cannot recognize a long QT when they see one. Heart Rhythm 2005;2:569–574.
- 69. ↑
Müntener CR, Müntener C, Kupper J, et al. Vigilance for veterinary medicinal products: reports of adverse reactions in the year 2016. Schweiz Arch Tierheilkd 2017;159:581–585.
- 70. ↑
Tjälve H. Adverse reactions to veterinary drugs reported in Sweden during 1991–1995. J Vet Pharmacol Ther 1997;20:105–110.
