RESUMO
AIMS: To compare the effect on mortality and length of hospital stay of propofol with that of sodium thiopentone for the management of dogs with status epilepticus (SE) and refractory status epilepticus (RSE). METHODS: In this cohort study, medical records of a veterinary referral clinic in Argentina were retrospectively searched for dogs that were hospitalised and required induction of therapeutic coma (TC) with either propofol or sodium thiopentone for the management of SE or RSE of any cause. A logistic regression model was performed to evaluate the association between the type of anaesthetic used and in-hospital mortality adjusting for the type of epilepsy (idiopathic, structural, or reactive). Kaplan-Meier estimated survival curves for the length of hospital stay by the type of anaesthetic drug were compared using the log-rank test (deaths were considered censored events). Cox proportional hazards regression was used to estimate hazard ratios for time to hospital discharge, unadjusted and adjusted for type of epilepsy. RESULTS: A total of 24 dogs with SE were included in the study: eight treated with propofol and 16 treated with sodium thiopentone. Four dogs treated with propofol (proportion = 0.50; 95% CI = 0.15-0.84), and eight treated with sodium thiopentone (proportion = 0.50; 95% CI = 0.50-0.74) died during hospitalisation. The median hospitalisation time was 43 (IQR 24-56) hours for dogs that were treated with propofol and 72 (IQR 64-96) hours for dogs that were treated with sodium thiopentone. There was no evidence of a difference in the median duration of TC in dogs treated with propofol (12 (IQR 8-24) hours) or with sodium thiopentone (12 (IQR 7.5-20) hours; p = 0.946). In the logistic regression model, no evidence of association between the anaesthetic protocol for the management of RSE and in-hospital mortality, adjusted for the type of epilepsy, was found (OR 1.09 (95% CI = 0.17-6.87); p = 0.925). Cox regression analysis revealed a difference in the time to hospital discharge, adjusted by the type of epilepsy, between treatment groups (HR = 0.05 (95% CI = 0.01-0.54); p = 0.013). CONCLUSIONS AND CLINICAL RELEVANCE: The time spent in hospital before discharge was longer in dogs with RSE treated with sodium thiopentone compared to those treated with propofol. However, as the sample size was very small, the results obtained in the present study should be analysed with caution. Further studies including a greater number of dogs are required.
Assuntos
Anestésicos , Doenças do Cão , Propofol , Estado Epiléptico , Cães , Animais , Tiopental/uso terapêutico , Tiopental/farmacologia , Propofol/uso terapêutico , Propofol/farmacologia , Estudos de Coortes , Estudos Retrospectivos , Estado Epiléptico/tratamento farmacológico , Estado Epiléptico/veterinária , Anestésicos/uso terapêutico , Sódio/uso terapêutico , Doenças do Cão/tratamento farmacológicoRESUMO
AIMS: To evaluate the echocardiographic variable tricuspid annular plane systolic excursion normalised to body weight (TAPSEnorm) as a predictor of fluid responsiveness in hospitalised dogs with haemodynamic and tissue perfusion alterations and to investigate the association of left ventricular internal diameter in diastole normalised to body weight (LVIDdN) and aortic velocity time integral (VTIAo) with TAPSEnorm. METHODS: A single-centre, prospective study was carried out in a cohort of spontaneously breathing dogs, hospitalised for any reason, with severe haemodynamic and tissue perfusion alterations. The echocardiographic variables TAPSEnorm, LVIDdN, and VTIAO were measured. A bolus of 30â mL/kg of lactated Ringer's solution was administered and then VTIAo was subsequently remeasured. Patients were classified as fluid responsive if VTIAo increased by ≥15% after fluid expansion, or non-responsive if VTIAo increased by <15% after fluid expansion. The area under the receiver operating characteristic (AUROC) curve was generated to evaluate the ability of TAPSE to predict fluid responsiveness. Simple regression models were used to assess the linear relationship between TAPSEnorm and LVIDdN or VTIAO. RESULTS: TAPSEnorm was lower in fluid responsive dogs (mean 0.57 (95% CI = 0.50-0.64) cm/kg) compared to non-responsive dogs (mean 0.76 (95% CI = 0.62-0.90) cm/kg). The AUROC for TAPSEnorm was 0.827 (95% CI = 0.65-1.00). The optimal cut-off point was 0.76 with sensitivity of 80 (95% CI = 28.4-99.5)% and specificity of 86.7 (95% CI = 69.3-99.2)%, positive predictive value of 50 (95% CI = 15.7-84.3)% and negative predictive value of 96.3 (95% CI = 81-99.9)%. A monotonic linear relationship was observed between TAPSEnorm and LVIDdN (p<0.001) and between TAPSEnorm and VTIAo (p=0.001). CONCLUSIONS AND CLINICAL RELEVANCE: TAPSEnorm could be useful in determining those dogs that are likely to respond to a fluid bolus from those that are likely to be non-responsive. Additionally, a positive linear association between the LVIDdN and the TAPSEnorm suggests that TAPSEnorm decreases at lower preload values. The present study results suggest that TAPSEnorm could be a valuable tool for evaluating blood volume status and fluid responsiveness in hospitalised dogs.Abbreviations: AUROC: Area under the receiver operating characteristic; CO: Cardiac output; ICC: Intraclass correlation coefficient; LVIDd: Left ventricular internal diameter in diastole; LVIDdN: Left ventricular internal diameter in diastole normalised to body weight; TAPSE: Tricuspid annular plane systolic excursion; TAPSEnorm: Tricuspid annular plane systolic excursion normalised to body weight; VTIAo: Aortic velocity time integral.