Introduction
Dexmedetomidine (DXM) is a sedative, muscular relaxant, and analgesic drug in common use in veterinary medicine.1–3 Its α-agonist effect is responsible for arterioconstriction, which is reflected clinically in a transiently increased arterial blood pressure.4,5 In cats, the resultant increase in left ventricular (LV) afterload is associated with enlargement of the cardiac silhouette on thoracic radiographs,6 and in combination with other drugs, it has been shown to alter LV chamber dimensions.4,5,7 Such changes have been investigated with drug combinations that include DXM in healthy cats sedated concurrently with midazolam and butorphanol,7 ketamine,7 or buprenorphine4 and in cats under isoflurane anesthesia.5 However, only limited information is available on the effects of DXM alone, followed by reversal with the α2-antagonist atipamezole (APM), on echocardiographic cardiac parameters.
An acute increase in LV afterload presents an opportunity to evaluate the dynamics of circulating cardiac biomarker concentrations in cats and their presumed return toward baseline after reversal of the drug that is causing the increase in afterload. Changes in circulating concentrations of cardiac biomarkers are associated with heart disease in cats,8,9 but neither the possible change in circulating concentrations triggered by an acute increase in cardiac work nor the possible reverse change in circulating concentrations caused by sudden elimination of such work is well understood in the cat.
This study aimed to evaluate the changes in certain echocardiographically measured cardiac dimensions and the changes in circulating cardiac biomarker concentrations in cats sedated with DXM, followed by reversal of DXM with APM. The null hypotheses were that in cats, echocardiographic observations and measurements are unchanged by administration of DXM and its reversal with APM and that circulating concentrations of cardiac troponin-I (cTn-I) and N-terminal pro-B–type natriuretic peptide (NT-proBNP) are not affected by DXM administration or APM reversal.
Materials and Methods
Animal selection
Privately owned domestic pet cats were volunteered by faculty, staff, and students in response to posters and email solicitation at the authors’ institution. The protocol was approved by the institutional review board. The power analysis for sample size determination, inclusion and exclusion criteria, and baseline tests have been described.6 Briefly, cats could be enrolled if they were > 12 months of age, had had no medical conditions that required overnight hospitalization in the preceding 6 months or treatment with medication in the last 30 days (except antiparasitics, vitamins, and nutraceuticals), had no known allergy or adverse reaction to α2-adrenergic agonists or other sedatives, had normal physical examination findings, and had informed consent provided in writing by the owner.
Cardiovascular evaluation
Participating cats underwent blood pressure measurement, blood sample collection, and diagnostic imaging at baseline (PRE) without sedation. The PRE assessment was followed by the experimental protocol on the same day (n = 9) or 1 to 6 days later (5). Tests performed PRE consisted of systolic arterial blood pressure measurement (Parks Doppler)9; a CBC, serum biochemistry profile, serum total thyroxine concentration, high-sensitivity measurement of serum cTn-I concentration (Architect Stat Troponin-I; Abbott Laboratories), and measurement of plasma NT-proBNP concentration (CardioPet; IDEXX Laboratories); a complete echocardiogram10; and 4-view thoracic radiographs. Inclusion criteria were CBC and serum biochemistry profile results that were within the laboratory’s reference interval or with abnormalities that were considered insignificant by the investigators; a serum total thyroxine concentration that was within the laboratory’s reference interval; mean of 4 systolic arterial blood pressure measurements < 180 mm Hg and no measurement > 180 mm; echocardiographic measurements within reference interval, including interventricular septal and LV free-wall thicknesses ≤ 5.5 mm in diastole in both M-mode and 2-D measurements and left atrial-to-aortic ratio (LA:Ao) < 1.5 measured at the end of the T wave on a concurrently recorded ECG; and thoracic radiographic findings interpreted as within normal limits. Echocardiographic measurements included LV M-mode measurements11 and 2-D LV wall measurements12; for the latter, diastolic interventricular septal thickness (2-D IVSd) was measured both at basal septal and midseptal levels, as previously described.12 Similarly, 2-D measurements of LV free-wall diastolic thickness (2-D LVFWd) were obtained.12 LA diameter was measured on the right-sided long-axis 4-chamber view at end systole13 and on short axis,14 and the LA area was measured on long axis.15 The LA:Ao valvar diameter ratio was calculated by use of images from a short-axis imaging plane in the first frame after Ao valve closure, as previously described.14 The LA area was calculated by tracing the endocardial border of the LA and applying the echocardiograph’s planimetry function (GE Logiq 7). All echocardiograms were performed by an American College of Veterinary Internal Medicine (ACVIM) board-certified cardiologist or an ACVIM cardiology resident under the direct supervision of an ACVIM board-certified cardiologist. All echocardiographic measurements were made in triplicate and averaged. These measurements were made by 1 cardiologist after images had been acquired for all 3 periods for a cat; the cardiologist was not blinded to the time period of each echocardiogram. When the instantaneous heart rate was variable (eg, during respiratory sinus arrhythmia), an effort was made to perform echocardiographic measurements on 3 beats with a similar preceding R-R interval to minimize the influence of ventricular loading. Valvar regurgitation was assessed subjectively with color-flow Doppler echocardiography and was recorded semiquantitatively by use of the following scale: none = 0, trace = 1, mild = 2, moderate = 3, and severe = 4. For each time period, a cumulative valvar regurgitation score (maximum total = 14 X 4 = 56) was calculated for each valve by summing the values for all cats for that valve. A cat’s heart rate for each time period was measured manually, in triplicate, from the ECG tracing on the LV M-mode image. The heart rhythm was assessed from the same ECG by an ACVIM board-certified cardiologist. For the purpose of comparisons between time periods, 6 ordered categories for heart rhythm on ECG were created. If there was variation between 2 rhythms on ECG during the echocardiogram, both rhythms were noted for that cat in that time period. A numerical score was assigned to each of the 6 rhythm categories: 1 = sinus bradycardia (≤ 110 beats/min), 2 = sinus bradycardia/normal sinus rhythm, 3 = respiratory sinus arrhythmia, 4 = normal sinus rhythm, 5 = normal sinus rhythm and sinus tachycardia, and 6 = sinus tachycardia (≥ 170 beats/min). LV chamber volume was determined by tracing the endocardial border and calculated using the Simpson method of discs from a right parasternal long-axis 4-chamber view.7,16 LV mass was calculated from the same view by tracing the epicardial border of the LV and calculating the volume using the Simpson method of discs, subtracting LV chamber volume, and then multiplying the result by the specific gravity of myocardium (1.05 g/mL).17 For measurement of cardiac biomarker concentrations, whole blood was placed in glass tubes containing tripotassium EDTA (1 aliquot) and no additive (1 aliquot), and both tubes were centrifuged at 2,012 X g within 30 minutes of collection. Harvested plasma was placed in plastic tubes containing a protease inhibitor and harvested serum in plain plastic tubes. The samples were refrigerated at –4 °C for < 12 hours, then transported on ice to be stored at –80 °C. This storage period lasted for 14 to 49 days (NT-proBNP) and 19 to 54 days (c-TnI), after which the samples were shipped by courier on dry ice for batch analysis at a central laboratory. Cats were excluded from further participation in the study if physical examination abnormalities were noted, echocardiographic results PRE exceeded reference intervals, echocardiographic results PRE revealed structural heart lesions, or the cats were not amenable to handling without sedation.
Protocol for assessing effects of DXM
Cats owners were instructed to withhold food for ≥ 12 hours prior to PRE. All cats then received DXM (Dexdomitor) IM at a dosage of 40 µg/kg per label instructions in the right epaxial muscles. Sedation was reported using an observational scale constructed from clinical observations (Supplementary Table S1). Clinical observations of sedation were performed every 5 minutes, and a score was assigned that most closely matched the observed findings. Tests were repeated during sedation (INTRA), defined as 15 minutes after DXM injection or 20 minutes after DXM injection if cats had a sedation score of 0 to 1 at 15 minutes after DXM injection. The procedures repeated INTRA were an echocardiogram, thoracic radiographs, and phlebotomy for measurement of serum cTn-I and plasma NT-proBNP concentrations. The duration of sedation, from DXM injection to reversal, depended on the time needed to perform these procedures. Once the procedures were complete, all cats received the reversal agent APM (Antisedan; 400 µg/kg) IM in the right epaxial muscles. Between 2 and 4 hours (2H POST) were allowed to pass following APM administration, during which time each cat rested in a cage. During this time, sedation again was assessed every 5 minutes, as described above, until a sedation score of 0 was achieved. At 2H POST, an echocardiogram, thoracic radiographs, and phlebotomy for biomarker analysis were repeated in the same manner described for INTRA, after which cats were discharged to their owners. Finally, cats were brought back for blood sample collection to assess circulating cTn-I and NT-proBNP concentrations the following day, approximately 24 hours later (24H POST).
Long-term follow-up
Long-term information on outcomes was sought by attempting to contact cat owners 8.5 to 9 years after each cat had participated in the study (Supplementary Table S2). A distinction was not made between spontaneous death and death by euthanasia. Data were collected for descriptive statistics only.
Statistical methods
The data constitute repeated measures across periods within each cat and were analyzed by linear mixed models with a fixed effect of period (Stata 16; StataCorp LLC). Biomarker concentrations, ECG data, color-flow Doppler echocardiographic valvar regurgitation data, and echocardiographic measurements were analyzed separately. Box-Cox analysis suggested whether transformation could improve the model assumptions of normality and homoscedasticity. Hypotheses about the structure of the within-cat residual errors were tested by likelihood ratio tests, and the most parsimonious error structure supported by the data was used. Estimated means or back-transformed estimated medians from the models are presented. Pairwise comparisons of the periods with the Holm adjustment for multiple comparisons were performed if the overall P value for period was significant. The Spearman rank correlation coefficient (rho) was used to evaluate the strength and direction of monotonic relationship between duration of sedation, defined as time in minutes from DXM administration to APM administration, and either cTn-I or NT-proBNP concentration at 2H POST and 24H POST. The scores of the ECG and color-flow Doppler echocardiographic valvar regurgitation variables were analyzed with the nonparametric Friedman test (P values obtained by simulation) as well as the Holm adjustment for pairwise comparisons. Results were considered statistically significant if P < 0.05.
Results
Demographic information and information regarding sedation
Of 29 volunteered cats, 15 were excluded due to noncooperation, leaving 14 that completed the study. Thirteen cats were domestic shorthairs and 1 cat was a domestic longhair. Nine cats were male, 5 were female, and all were neutered. The median age was 5 years (range, 2 to 11 years), and the median body weight was 5.2 kg (range, 3.5 to 7.8 kg). Thirteen of 14 cats had a sedation score ≥ 4 at 12 to 16 minutes after DXM administration; cat No. 2 had a sedation score of 0 at 15 and 20 minutes postadministration of DXM. For those cats that vomited and for which the time between DXM administration and vomiting was recorded, the median time was 5 minutes (range, 3 to 20 minutes; n = 6). The duration of sedation was 62 ± 23 minutes (range, 35 to 111 minutes).
ECG and 2-D and M-mode echocardiography
Heart rate and rhythm were significantly different among time periods (Tables 1 and 2). Echocardiographic data are shown (Tables 1 and 3). Certain measurements of IVSd and LVFWd thickness were significantly but minimally different across time periods. M-mode IVSd (P = 0.15), 2-D IVSd (midseptal; P = 0.5), and 2-D LVFWd (P = 0.4) did not change significantly. The M-mode LV systolic diameter (LVIDs) was significantly increased but the M-mode diastolic LV diameter (LVIDd) was not (P = 0.35). The LA area measured on the long axis (P = 0.22), LA:Ao (P = 0.84), and LV mass (P = 0.35) did not change significantly. The long-axis LA diameter was not significantly different for the change across time overall (P = 0.05) but was greater INTRA than PRE.
Descriptive statistics for selected echocardiographic variables, by time periods.
| Variables | Period | |||||||
|---|---|---|---|---|---|---|---|---|
| PRE | INTRA | 2H POST | 24H POST | |||||
| Median | IQR | Median | IQR | Median | IQR | Median | IQR | |
| HR (beats/min) | 188.167 | 53.667 | 95.500 | 26.667 | 162.000 | 29.333 | — | — |
| M IVSd (mm) | 4.100 | 0.460 | 3.870 | 0.730 | 3.920 | 0.900 | — | — |
| M LVIDd (mm) | 14.865 | 3.500 | 16.030 | 3.400 | 15.685 | 2.430 | — | — |
| M LVFWd (mm) | 4.330 | 0.600 | 3.935 | 0.770 | 3.850 | 0.800 | — | — |
| M LVIDs (mm) | 8.733 | 2.367 | 11.650 | 4.133 | 10.467 | 1.933 | — | — |
| 2-D IVSd-LAX1 (mm) | 4.000 | 0.700 | 3.950 | 0.930 | 3.700 | 0.460 | — | — |
| 2-D IVSd-LAX2 (mm) | 3.785 | 0.630 | 3.965 | 1.130 | 3.515 | 1.200 | — | — |
| 2-D LVFWd (mm) | 3.970 | 0.570 | 3.780 | 0.670 | 3.900 | 0.670 | — | — |
| 2-D LVEDV (mL) | 2.795 | 1.160 | 3.495 | 1.300 | 2.810 | 1.020 | — | — |
| 2-D LVESV (mL) | 0.597 | 0.633 | 1.900 | 0.740 | 1.158 | 0.523 | — | — |
| 2-D LVEF (%) | 77.5 | 15.0 | 48.0 | 12.0 | 67.2 | 13.0 | — | — |
| LV mass (g) | 5.255 | 1.520 | 5.625 | 1.890 | 5.405 | 1.070 | — | — |
| LA:Ao | 1.253 | 0.140 | 1.152 | 0.413 | 1.223 | 0.117 | — | — |
| 2-D LAD_lax (mm) | 12.767 | 2.333 | 14.150 | 1.733 | 13.367 | 2.133 | — | — |
| LAA_lax (cm2) | 1.895 | 0.730 | 1.960 | 0.680 | 1.890 | 0.710 | — | — |
| 2-D LAD_sax (mm) | 8.817 | 1.833 | 10.767 | 0.933 | 9.933 | 1.733 | — | — |
| RVOT (m/s) | 0.850 | 0.240 | 0.460 | 0.060 | 0.560 | 0.120 | — | — |
| LVOT (m/s) | 0.905 | 0.140 | 0.530 | 0.140 | 0.770 | 0.080 | — | — |
| cTn-I (ng/mL) | 0.016 | 0.010 | 0.025 | 0.028 | 0.059 | 0.190 | 0.045 | 0.040 |
| NT-proBNP (pmol/L) | 49.000 | 40.000 | 68.000 | 87.000 | 63.500 | 94.000 | 84.500 | 8.000 |
— = Not available. 24H POST = 24 hours after DXM administration. 2-D_IVSd-LAX1 = 2-D interventricular septal thickness in diastole (long axis segment 1). 2-D_IVSd-LAX2 = 2-D interventricular septal thickness in diastole (segment 2). 2-D_LAD_lax = 2-D left atrial diameter in long axis. 2-D_LAD_sax = 2-D left atrial diameter in short axis. 2-D_LVFWd = 2-D left ventricular (LV) free-wall thickness in diastole. 2-D_LVEDV = 2-D LV volume in diastole. 2-D_LVESV = 2-D LV volume in systole. 2-D_LVEF = 2-D LV ejection fraction. 2H POST = 2 to 4 hours after APM administration. cTn-I = Serum concentration of cardiac troponin-I. HR = Heart rate. INTRA = Measurements obtained during sedation. LAA_lax = Left atrial area in long axis. LA:Ao = Left atrial-to-aortic ratio. LVOT = Doppler echocardiography–derived peak velocity of blood flow in the LV outflow tract. M_IVSd = M-mode–derived interventricular septal thickness in diastole. M_LVIDd = M-mode–derived LV internal diameter in diastole. M_LVFWd = M-mode–derived LV free-wall thickness in diastole. M_LVIDs = M-mode–derived LV internal diameter in systole. NT-proBNP = Plasma concentration of amino-terminal pro-B–type natriuretic peptide. PRE = Measurements obtained at baseline. RVOT = Doppler echocardiography–derived peak velocity of blood flow in the right ventricular outflow tract.
Descriptive statistics for ECG rhythm.
| Time period/ heart rhythm | No. of cats with ECG rhythms in each time period | |||||
|---|---|---|---|---|---|---|
| SB | SB/NSR | RSA | NSR | NSR/ST | ST | |
| PRE | 0 | 0 | 0 | 3 | 3 | 8 |
| INTRA | 3 | 3* | 6 | 2 | 0 | 0 |
| 2H POST | 0 | 0 | 2 | 3 | 7 | 2 |
ST = Sinus tachycardia.
Includes 1 cat with normal sinus rhythm (NSR), respiratory sinus arrhythmia (RSA), and sinus bradycardia (SB) with premature atrial complexes and ventricular escape beats.
The distribution of values is significantly different from PRE to INTRA (P < 0.001), INTRA to 2H POST (P < 0.001), and PRE to 2H POST (P = 0.022).
See Table 1 for remainder of key.
Estimated means (medians for variables indicated with * or **), 95% confidence intervals, and pairwise comparisons from linear mixed model analysis.
| Variable | Period | ||||
|---|---|---|---|---|---|
| PRE (95%CI) | INTRA (95%CI) | 2H POST (95%CI) | 24H POST (95%CI) | P value# | |
| Heart rate* (beats/min) | 186a (169–204) | 100c (91–110) | 152b (139–167) | — | < 0.001 |
| M-mode LVFWd (mm) | 4.31a (3.97–4.64) | 3.89b (3.56–4.22) | 3.84b (3.51–4.17) | — | 0.020 |
| M-mode LVIDs (mm) | 8.94c (7.94–9.94) | 11.92a (10.92–12.92) | 10.53b (9.53–11.53) | — | < 0.001 |
| 2-D IVSd-LAX1 (mm) | 4.16a (3.89–4.43) | 4.02a,b (3.76 – 4.29) | 3.81b (3.54–4.07) | — | 0.015 |
| 2-D LAD_lax (mm) | 13.46b (12.48–14.44) | 14.59a (13.61–15.57) | 14.03a,b (13.05–15.01) | — | 0.053 |
| 2-D LAD_sax* (mm) | 9.18c (8.61–9.79) | 11.05a (10.37–11.79) | 10.01b (9.39–10.67) | — | < 0.001 |
| 2-D LVEDV* (mL) | 3.08b(2.67–3.55) | 3.52a (3.06–4.06) | 3.01b(2.61–3.46) | — | < 0.001 |
| 2-D LVESV* (mL) | 0.66c (0.52–0 .84) | 1.75a(1.3–2.22) | 1.05b(0.83–1.33) | — | < 0.0001 |
| 2-D LVEF (%) | 76.71a (71.34–82.09) | 49.00c (43.63–54.37) | 63.41b (58.04–68.79) | — | < 0.001 |
| Doppler RVOT* Vmax (m/s) | 0.88a (0.82–0.95) | 0.44c (0.41–0.48) | 0.60b (0.56–0.65) | — | < 0.001 |
| Doppler LVOT Vmax (m/s) | 0.94a (0.89–0.99) | 0.52c (0.47–0.57) | 0.79b(0.74–0.84) | — | < 0.001 |
| cTn-I* (ng/mL) | 0.020c(0.014–0.029) | 0.028b,c(0.014–0.058) | 0.073a(0.036–0.146) | 0.042a,b (0.023–0.076) | < 0.001 |
| NT-proBNP** (pmol/L) | 42.2b (18.2–76.0) | 61.1a,b (31.3–100.9) | 73.7a (40.5–116.9) | 72.6a (39.6–115.5) | 0.007 |
Vmax = Peak velocity.
The estimated means or medians sharing the same superscript are not significantly different at the 5% level.
Natural log transformed.
Square root transformed.
Overall P value for comparison of the 3 periods.
See Table 1 for remainder of key.
Color-flow Doppler echocardiography
The presence of valvar regurgitation at each time period is shown (Table 4). A manyfold increase in the number of cats with an onset or increase in valvar regurgitation and a significant increase in cumulative valvar regurgitation score were noted for all valves from PRE to INTRA. At 2H POST, the valvar regurgitation score remained different from PRE for mitral regurgitation but had returned to not being significantly different from PRE for the other 3 valves. The difference in valvar regurgitation score across the 3 time periods was significant for all 4 valves. Neither 2-D nor Doppler echocardiographic evidence of dynamic right ventricular or LV outflow obstruction was apparent in any cat at any time period. Spectral Doppler echocardiographic transvalvar Ao and pulmonary velocities had 1 missing value for INTRA and for 2H POST in each data set; the remaining values were significantly different among time periods for both Ao and pulmonary valves (Table 3). Spectral Doppler echocardiographic transmitral E and A waves were either fused (n = 8) or not measurable or recorded (4) for at least 1 time period, precluding analysis.
Analysis of color-flow Doppler echocardiographic results.
| Variable | Period | p value# | |||
|---|---|---|---|---|---|
| PRE | INTRA | 2H POST | |||
| MR | No. of cats affected* | 1 | 14 | 6 | |
| Total score** | 1c | 26a | 8b | < 0.001 | |
| TR | No. of cats affected* | 4 | 10 | 6 | |
| Total score** | 6a | 14a | 8a | 0.049 | |
| AR | No. of cats affected* | 2 | 12 | 4 | |
| Total score** | 2b | 21a | 4b | < 0.001 | |
| PR | No. of cats affected* | 0 | 8 | 2 | |
| Total score** | 0b | 9a | 2a,b | 0.001 | |
AR = Aortic valvar regurgitation. MR = Mitral valvar regurgitation. PI = Pulmonary valvar regurgitation. TR = Tricuspid valvar regurgitation.
Out of 14 cats.
Out of maximal score of 56.
Overall P value for comparison of the 3 periods.
Total scores sharing the same superscript correspond to distributions of scores that are not significantly different at the 5% level.
See Table 1 for remainder of key.
Plasma cardiac biomarker concentrations
Circulating cardiac biomarker concentrations changed significantly over time (Table 1; Figure 1). Both serum cTn-I concentration (P = 0.0004) and plasma NT-proBNP concentration (P = 0.007) increased significantly at 2H POST compared to PRE, but not at INTRA compared to PRE. Values at 24H POST remained significantly higher than values at PRE for both biomarkers. Across the 4 time periods, the median difference between maximal and minimal serum cTn-I concentration was 0.053 ng/mL, and the median difference between maximal and minimal plasma NT-proBNP concentration was 32 pmol/L. One cat had PRE, INTRA, 2H POST, and 24H POST serum cTn-I concentrations of 0.073, 0.832, 1.416, and 0.46 ng/mL, respectively, representing a difference of 1.343 ng/mL. Six cats had plasma NT-proBNP concentrations ≥ 100 pmol/L at 1 or more time periods. A significant (P = 0.009), moderate positive (rho = 0.67) association existed between duration of sedation and serum cTn-I concentration at 24H POST, and a weakly significant (P = 0.048), moderate positive (rho = 0.54) association existed between duration of sedation and serum cTn-I concentration at 2H POST. No significant association existed between duration of sedation and plasma NT-proBNP concentration at either 2H POST (rho = –0.13; P = 0.659) or 24H POST (rho = –0.40; P = 0.158).
Circulating concentrations of cardiac troponin-I (cTn-I; top panel) and N-terminal pro-B–type natriuretic peptide (NT-proBNP; bottom panel) at each of 4 time periods for the 14 cats studied. PRE = Measurements obtained at baseline. INTRA = Measurements obtained during sedation. 2H POST = 2 to 4 hours after APM administraton. 24H POST = 24 hours after DXM administration.
Citation: Journal of the American Veterinary Medical Association 260, 8; 10.2460/javma.21.06.0299
Circulating concentrations of cardiac troponin-I (cTn-I; top panel) and N-terminal pro-B–type natriuretic peptide (NT-proBNP; bottom panel) at each of 4 time periods for the 14 cats studied. PRE = Measurements obtained at baseline. INTRA = Measurements obtained during sedation. 2H POST = 2 to 4 hours after APM administraton. 24H POST = 24 hours after DXM administration.
Citation: Journal of the American Veterinary Medical Association 260, 8; 10.2460/javma.21.06.0299
Circulating concentrations of cardiac troponin-I (cTn-I; top panel) and N-terminal pro-B–type natriuretic peptide (NT-proBNP; bottom panel) at each of 4 time periods for the 14 cats studied. PRE = Measurements obtained at baseline. INTRA = Measurements obtained during sedation. 2H POST = 2 to 4 hours after APM administraton. 24H POST = 24 hours after DXM administration.
Citation: Journal of the American Veterinary Medical Association 260, 8; 10.2460/javma.21.06.0299
Externally observable changes with APM administration
The return to sedation score of 0 (unsedated appearance) occurred at a median of 25 minutes after APM administration (range, 10 to 25 minutes; n = 12; 1 outlier with a sedation score of 0 at 15 and 20 minutes and sedation score of 2 at 1 hour [1 not recorded]).
Long-term follow-up
Cats owners (14/14 [100%]) responded to requests for information pertaining to long-term outcomes 8.8 to 8.9 years after the cats’ participation in the study (Supplementary Table S2). Seven (50%) cats were alive, with a median age of 12 (range, 10 to 15) years. No cat was known to have died of heart disease.
Discussion
This study of DXM in cats confirmed certain cardiovascular effects noted previously in response to administration of DXM-containing drug combinations to cats. These effects included a marked decrease in heart rate and associated rhythm,4,7 minor increase in M-mode LVIDd,4 marked increase in LVIDs,4,7 increase in LV end-systolic volume,7 decrease in LV ejection fraction,7 lack of significant change in M-mode IVSd,4,7 marked reductions in peak right ventricular and LV outflow tract velocities,4 and onset of valvar regurgitation.5
New information presented in this study included certain changes in cardiac dimensions as assessed using M-mode, 2-D, and planimetric echocardiographic methods. Of foremost interest was whether DXM sedation could temporarily reduce the thickness of LV walls in a way that could mask hypertrophic cardiomyopathy. The results of the present study indicate such type II error is unlikely. For IVSd, a significant difference was apparent with 2-D in 1 septal segment on long axis (LAX1) but not in the other septal segment (LAX2) nor on M-mode. Furthermore, the difference in the 2-D septal segment thickness that measured lower INTRA compared to PRE was statistically significant but of a magnitude that is clinically insignificant (median, 0.14 mm). Similarly, M-mode LVFWd was significantly lower INTRA compared to PRE (mean difference, 0.42 mm) but no significant difference was apparent for LVFWd on 2-D measurements. These results show that DXM 40 µg/kg IM can change LV wall thickness measurements in a way that affects certain segments measured using certain echocardiographic methods but not others, and that this effect is of limited magnitude in those segments when it occurs.
Changes in LA size caused by DXM administration were mixed. The 2-D LA diameter on LAX increased from PRE to INTRA (difference of means, 1.13 mm) as did the 2-D LA diameter on short axis (difference of means, 1.87 mm), but the 2-D LA:Ao did not. The lack of change in LA:Ao could be explained by other factors, such as a concomitant increase in Ao diameter or poor sensitivity of this imaging plane for LA enlargement. LA enlargement has been noted in cats sedated with DXM in 1 study4 but not in 2 others.5,7 The present study involved measuring LA dimensions in 3 ways: linearly (short- and long-axis diameters) and using planimetry (long-axis surface area). The increase in short-axis LA diameter from PRE to INTRA and lack of significant difference in long-axis LA area between time periods suggests that a DXM-associated change in LA size is likely to be of limited magnitude and may not be detectable in more than 1 imaging plane. Changes in absolute LA diameter of the magnitude found in the present study are unlikely to alter the interpretation of a cat’s echocardiogram. Changes in LA size have been reported with other cardiovascular drug treatments in cats.18,19
Also of interest were the varying degrees of reversal of echocardiographic changes of DXM-sedated cats when given APM. Of the variables that changed significantly from PRE to INTRA, some returned to being not significantly different from PRE at 2H POST, indicating APM-induced reversal of DXM changes: 2-D left atrial diameter-LAX and 2-D LV volume in diastole. Others remained significantly different from PRE at 2H POST, either with (indicating partial reversal; heart rate, M-mode LVIDs, 2-D left atrial diameter in short axis, LV volume in systole, LV ejection fraction, Doppler echocardiography–derived peak velocity of blood flow in the right ventricular outflow tract, and Doppler echocardiography–derived peak velocity in the LV outflow tract) or without (indicating no significant reversal; M-mode LVFWd and 2-D IVSd-LAX1) being significantly different from INTRA. Together, these findings demonstrate that echocardiographic changes in cats given DXM return toward baseline measurements with varying rate and extent for 2 or more hours after reversal with APM.
Certain results can be explained by echocardiographic methods. An increase in LV volume in diastole when calculated with the Simpson method of discs has not been reported previously in association with DXM administration in cats, and this stands in contrast to the lack of change in LVIDd. A plausible explanation is the greater precision associated with a three-dimensional measurement rather than a linear measurement. A clinical implication could be that a cat with an increased LVIDd during sedation with DXM is more likely to have this as an intrinsic finding (eg, due to pathologic LV dilation from eccentric LV hypertrophy, since DXM does not alter LVIDd according to the present results) than as an effect of DXM.
The lack of difference in measured LV mass across time periods is intuitive. However, a change in loading conditions has been observed to alter echocardiographically calculated myocardial mass in humans undergoing treatments that alter intravascular volume (hemodialysis20 and furosemide administration21). No such effect was observed in the cats of this study.
Valvar regurgitation occurred with DXM administration in many cats, including mitral regurgitation in 14 of 14 cats when only 1 of 14 cats had it PRE. The higher occurrence of mitral valvar regurgitation compared to Ao valvar regurgitation can be explained by the higher ventricular pressure generated during systole in the face of an acute increase in afterload caused by an α2-agonist drug such as DXM. The occurrence of valvar regurgitation that was greater for left-sided cardiac valves than right-sided cardiac valves can be explained by the higher pressures generated in the left heart and aorta compared to those in the right heart and pulmonary artery, particularly under the effect of an α-agonist such as DXM. A reduction in the number of insufficient valves and valvar regurgitation score at 2H POST compared to INTRA suggests that valvar regurgitation is short-lived, but can persist as a DXM-related effect after APM has been administered and a cat is no longer visibly sedated. The importance of these findings is to recognize that valvar regurgitation is a widespread phenomenon in cats that receive DXM at the labeled dosage, should not be attributed to primary valvar disease, and is at least partially reversible within 2 to 4 hours of APM reversal. This finding may have applications in other species of cats.22
Circulating concentrations of both biomarkers were significantly higher 2H POST compared to PRE, and both remained significantly higher 24H POST compared to PRE. This represents an important confounder to be considered if measuring circulating cardiac biomarker concentrations in cats that are sedated with DXM or have been sedated with DXM in the 24 hours preceding blood sampling. A comparison of cats with occult cardiomyopathy to normal cats identified that a plasma NT-proBNP concentration > 46 pmol/L distinguished normal from occult cardiomyopathy with 91.2% specificity and 85.8% sensitivity, and that a > 99 pmol/L cutoff was 100% specific and 70.8% sensitive.9 In the present study, DXM increased the plasma NT-proBNP concentration from < 46 pmol/L to > 46 pmol/L in 3 of 14 cats and from < 99 pmol/L to > 99 pmol/L in 4 of 14 cats, at 1 or more time periods. Since the cats in the present study did not have cardiomyopathy, other cats with these results may not have cardiomyopathy if their blood sample was drawn within 24 hours of DXM administration. The increases in cardiac biomarker concentrations noted in the present study could be explained by a duration of DXM effects on the myocardium that is longer than the sedative effects, by delayed renal clearance of biomarker molecules if such clearance is reduced by DXM administration, or by other factors.
Complete resolution of visible sedation was apparent in 12 of 13 cats within 25 minutes of APM administration; then 2H POST echocardiograms and blood samples for measuring biomarker concentrations were obtained 2 to 4 hours after APM administration. Since significant differences existed between PRE and 2H POST measurements for several echocardiographic variables and both biomarkers, this study demonstrates that in most cats, visible sedation due to DXM fully resolves before many cardiovascular effects do.
A methodological consideration is the dosage of DXM administered to cats (40 µg/kg). This is the dosage that is approved by the USDA for use in cats,23 which was the basis for selecting it for this study. However, DXM routinely is administered at much lower dosages clinically (1 to 5 µg/kg). Whether cardiovascular effects that are comparable to the results of this study occur at lower dosages is a question that merits further investigation given the present findings. One echocardiographic study5 of cats IM treated with DXM 5 µg/kg identified a significant increase in LVIDs, compared with results at baseline, but the duration of the change was brief and the effect on LVIDs was no longer apparent even while the cats received an IV infusion of DXM at 1 µg/kg/min for 60 minutes.5
The results of the present study indicate that certain cats receiving 40 µg/kg of DXM IM may experience subclinical myocardial changes, including a substantial release of NT-proBNP and cTn-I (Figure 1). Two cats (No. 5 and No. 11) experienced marked increases in circulating NT-proBNP and cTn-I concentrations from PRE through INTRA to 2H POST, with a partial reduction in serum cTn-I concentration at 24H POST and a persistent elevation in plasma NT-proBNP concentration. These findings suggest that these cats experienced transient myocardial changes in association with DXM administration, although ventricular arrhythmias and physical manifestations of a cardiac disorder were not apparent in these or any other cats. Thus, DXM administered IM at 40 µg/kg may induce substantial, negative, transient myocardial effects that are detectable with NT-proBNP and high-sensitivity cTn-I analysis in a small number of healthy cats.
There was an association between serum cTn-I concentration at 24H POST and the duration of sedation, from DXM administration to APM reversal. This finding suggested that there are cardiovascular benefits to reversing DXM with APM when the need for sedation is over, rather than letting the effects of DXM resolve spontaneously. Together, the results involving cTn-I offer evidence to support a reduction in the labeled dosage of DXM in cats. These findings might not have been suspected otherwise, nor these conclusions reached, because no physical or ECG abnormalities were noted during the study that raised cardiovascular concerns.
In 1 cat, the sedation score initially did not change after DXM administration. An important consideration might be that the cat did not receive the DXM (ie, a technical error with the injection). This is considered an unlikely possibility because the cat vomited 20 minutes after DXM administration, the cat’s INTRA heart rate was 82 beats/min compared to its PRE heart rate of 157 beats/min, and the cat’s sedation score was 2 at 1 hour after DXM administration. This cat was considered an outlier in its response to DXM administration. This was a different cat from the 2 that had a high serum cTn-I concentration 2H POST.
Long-term follow-up and survival assessments are not widely reported for feline cardiac biomarker studies. The fact that no cat in this study was known to have developed clinical signs of heart disease nor to have died of a known cardiac cause, after follow-up > 8 years, is consistent with earlier large-scale case series of long-term survival in cats.24,25 It is intriguing that the cat with high serum cTn-I concentration at INTRA, 2H POST, and 24H POST is also the cat that developed HOCM 7 years later and the only cat to receive cardiovascular medications (cat No. 11). Further exploration of the role of biomarkers in detecting heart disease in cats prior to the appearance of echocardiographic changes is warranted.
Limitations include the fact that cats were excluded if they were not cooperative, which biased this study sample in favor of well-behaved cats; whether this bias affected the results is unknown. The power analysis that generated the sample size was oriented to thoracic radiographic findings, and whether or not adequate power would exist for echocardiographic and cardiac biomarker results was not known prior to performing the study. The cutoff of < 180 mm Hg for systolic blood pressure as an entry criterion could have allowed mildly or moderately hypertensive cats to be enrolled, although no cat showed evidence of target organ damage. The cardiologist performing echocardiographic measurements was not blinded to the time period, which could have introduced bias when echocardiographic measurements were made; however, the visible difference in heart rate between INTRA and both PRE and 2H POST could have made such blinding imperfect had it been applied. External influences such as the stress of travel and handling could have contributed to the increase in circulating cardiac biomarker concentrations over time. Within the defined and reported time periods, variation existed in exact timing and duration of events.
This study showed that DXM 40 µg/kg IM produces effective sedation in most cats and that this sedation is accompanied by cardiovascular changes that can outlast the visible effect of sedation after DXM has been reversed using APM. The echocardiographic changes are those that would be expected with an acute increase in afterload, often including transient valvar regurgitation. No DXM-associated change would be expected to cause pseudonormalization of LV wall thickening in a cat with hypertrophic cardiomyopathy because IVSd and LVFWd were not significantly and consistently lower INTRA or 2H POST compared to PRE. Certain echocardiographic changes begin to regress promptly after reversal with APM, whereas others last for at least 2 to 4 hours. Increases in circulating cardiac biomarker concentrations are significant and can last for at least 24 hours after DXM administration and reversal with APM. These findings may help veterinarians more accurately differentiate between expected drug effects and spontaneous cardiac disease in cats receiving DXM.
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org
Acknowledgments
Funded by grants from Zoetis (Pfizer Animal Health), the Atlantic Veterinary College Research Fund, and IDEXX Laboratories.
The investigators met with the study sponsors prior to recruiting animals. It was agreed that the investigators retained sole ownership of data and had sole control of the study protocol. One sponsor suggested adding an evaluation after reversal of DXM with atipamezole, where the study protocol originally evaluated cats only PRE and INTRA. The investigators agreed with making this change and added 2H POST and 24H POST time periods to the protocol a priori. The investigators and sponsors agreed that the investigators retained an unrestricted and independent right to publish.
The authors thank Dr. Andrea Matthews for input on study design and Elaine Reveler for technical support.
References
- 1. ↑
Ansah OB, Raekallio M, Vainio O. Comparison of three doses of dexmedetomidine with medetomidine in cats following intramuscular administration. J Vet Pharmacol Ther. 1998;21(5):380–387.
- 2.
Martinez EA. Anesthetic agents. In: Booth DM, ed. Small Animal Clinical Pharmacology and Therapeutics. 2nd ed. Elsevier; 2012:887–893.
- 3. ↑
McSweeney PM, Martin DD, Ramsey DS, McKusick BC. Clinical efficacy and safety of dexmedetomidine used as a preanesthetic prior to general anesthesia in cats. J Am Vet Med Assoc. 2012;240(4):404–412.
- 4. ↑
Johard E, Tidholm A, Ljungvall I, Häggström J, Höglund K. Effects of sedation with dexmedetomidine and buprenorphine on echocardiographic variables, blood pressure and heart rate in healthy cats. J Feline Med Surg. 2018;20(6):554–562.
- 5. ↑
Carvalho ER, Champion T, Ambrosini F, da Silva GA, Freitas GC, D’Otaviano de Castro Vilani RG. Dexmedetomidine low dose followed by constant rate infusion and antagonism by atipamezole in isoflurane-anesthetized cats: an echocardiographic study. Vet Anaesth Analg. 2019;46(1):43–54.
- 6. ↑
Zwicker LA, Matthews AR, Côté E, Andersen E. The effect of dexmedetomidine on radiographic cardiac silhouette size in healthy cats. Vet Radiol Ultrasound. 2016;57(3):230–236.
- 7. ↑
Biermann K, Hungerbühler S, Mischke R, Kästner SBR. Sedative, cardiovascular, haematologic and biochemical effects of four different drug combinations administered intramuscularly in cats. Vet Anaesth Analg. 2012;39(2):137–150.
- 8. ↑
Hori Y, Iguchi M, Heishima Y, et al. Diagnostic utility of cardiac troponin I in cats with hypertrophic cardiomyopathy. J Vet Intern Med. 2018;32(3):922–929.
- 9. ↑
Fox PR, Rush JE, Reynolds CA, et al. Multicenter evaluation of plasma N-terminal probrain natriuretic peptide (NT-pro BNP) as a biochemical screening test for asymptomatic (occult) cardiomyopathy in cats. J Vet Intern Med. 2011;25(5):1010–1016.
- 10. ↑
Thomas WP, Gaber CE, Jacobs GJ, et al. Recommendations for standards in transthoracic two-dimensional echocardiography in the dog and cat. J Vet Intern Med. 1993;7(4):247–252.
- 11. ↑
Jacobs G, Knight DH. M-mode echocardiographic measurements in nonanesthetized healthy cats: effects of body weight, heart rate, and other variables. Am J Vet Res. 1985;46(8):1705–1711.
- 12. ↑
Wagner T, Fuentes VL, Payne JR, McDermott N, Brodbelt D. Comparison of auscultatory and echocardiographic findings in healthy adult cats. J Vet Cardiol. 2010;12(3):171–182.
- 13. ↑
Maerz I, Schober K, Oechtering G. Echocardiographic measurement of left atrial dimension in healthy cats and cats with left ventricular hypertrophy. Tierarztl Prax Ausg K Klientiere Heimtiere. 2006;34(5):331–340.
- 14. ↑
Abbott JA, MacLean HN. Two-dimensional echocardiographic assessment of the feline left atrium. J Vet Intern Med. 2006;20(1):111–119.
- 15. ↑
Linney CJ, Dukes-McEwan J, Stephenson HM, López-Alvarez J, Fonfara S. Left atrial size, atrial function and left ventricular diastolic function in cats with hypertrophic cardiomyopathy. J Small Anim Pract. 2014;55(4):198–206.
- 16. ↑
Lang RM, Badano LP, Mor-Avi V, et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr. 2015;28(1):1–39.e14. doi:10.1016/j.echo.2014.10.003
- 17. ↑
MacDonald KA, Kittleson MD, Reed T, Larson R, Kass P, Wisner ER. Quantification of left ventricular mass using cardiac magnetic resonance imaging compared with echocardiography in domestic cats. Vet Radiol Ultrasound. 2005;46(3):192–199.
- 18. ↑
Campbell FE, Kittleson MD. The effect of hydration status on the echocardiographic measurements of normal cats. J Vet Intern Med. 2007;21(5):1008–1015.
- 19. ↑
Kochie SL, Schober KE, Rhinehart J, et al. Effects of pimobendan on left atrial transport function in cats. J Vet Intern Med. 2021;35(1):10–21.
- 20. ↑
Martin LC, Barretti P, Cornejo IV, et al. Influence of fluid volume variations on the calculated value of the left ventricular mass measured by echocardiogram in patients submitted to hemodialysis. Ren Fail. 2003;25(1):43–53.
- 21. ↑
Prisant LM, Kleinman DJ, Carr AA, Bottini PB, Gross CM. Assessment of echocardiographic left ventricular mass before and after acute volume depletion. Am J Hypertens. 1994;7(5):425–428.
- 22. ↑
Chai N, Petit T, Kohl M, et al. Prevalence of valvular regurgitations in clinically healthy captive leopards and cheetahs: a prospective study from the Wildlife Cardiology (WLC) Group (2008–2013). J Zoo Wildl Med. 2015;46(3):526–533.
- 23. ↑
Dexdomitor. Package insert. Zoetis; 2015. Accessed June 11, 2021. https://www.zoetisus.com/products/dogs/dexdomitor-01/doc/DEXDOMITOR_PI.pdf
- 24. ↑
Fox PR, Keene BW, Lamb K, et al. International collaborative study to assess cardiovascular risk and evaluate long-term health in cats with preclinical hypertrophic cardiomyopathy and apparently healthy cats: The REVEAL Study. (Erratum in: J Vet Intern Med. 2018;32(6):2310. doi:10.1111/jvim.15285). J Vet Intern Med. 2018;32(3):930–943.
- 25. ↑
Fox PR, Keene BW, Lamb K, et al. Long-term incidence and risk of noncardiovascular and all-cause mortality in apparently healthy cats and cats with preclinical hypertrophic cardiomyopathy. J Vet Intern Med. 2019;33(6):2572–2586.
