Responsividade, confiabilidade e interferência da presença de um observador humano ao se avaliar a dor pela expressão facial em coelhos
Carregando...
Data
2022-01-21
Autores
Pinho, Renata Haddad [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
Nas últimas décadas têm-se empregado escalas faciais como ferramentas para avaliar a dor em diversas espécies. Esse estudo objetivou avaliar a responsividade, confiabilidade e interferência de um observador humano nos escores da RbtGS (Rabbit Grimace Scale) por vídeo filmagens em coelhos submetidos à cirurgia ortopédica. Utilizaram-se os registros de 28 coelhos filmados como parte de outro estudo em quatro momentos: basal (24 horas antes da cirugia); dor (1 hora após a recuperação anestésica); analgesia (3 horas após o resgate analgésico e 4 horas após a recuperação anestésica) e 24h (24 horas após a recuperação anestésica) na presença ou na ausência do observador. A RbtGS foi pontuada de forma encoberta e aleatória por três avaliadores. Avaliou-se a responsividade comparandose os quatro momentos pelo teste de Friedman, seguido de Dunn. A interferência da presença ou ausência do observador em cada momento foi avaliada pelo teste de Wilcoxon. A confiabilidade intra e interobservador foi determinada pelo coeficiente de correlação intraclasse (ICC). A escala foi responsiva, pois os escores aumentaram após o procedimento cirúrgico e reduziram após 24 horas. A presença do observador reduziu os escores da RbtGS no momento dor (presença: 0,75; 0-1,75, ausência: 1; 0-2) e aumentou nos momentos basal (presença: 0,2; 0-2, ausência:0; 0-2) e 24h (presença: 0,33; 0-1,75, ausência: 0,2; 0-1,5). A confiabilidade intra e interobservador variou de moderada a muito boa (ICC > 0,49). A RbtGS apresentou responsividade, repetibilidade e reprodutibilidade na avaliação por vídeo filmagens em coelhos submetidos à cirurgia ortopédica. Na presença da observadora humana familiar aos coelhos, subestimou-se os escores de dor no momento de maior dor e superestimou-se a dor nos momentos de pouca ou nenhuma dor, sugerindo-se que as avaliações sejam realizadas remotamente. O número significante de dados perdidos é uma limitação importante da RbtGS.
In the last decades, facial pain scales have been used as tools to assess pain in several species. The goal of this study was to evaluate the responsiveness, the intra- and inter-observer reliability and the influence of a human observer on the scores of the RbtGS (Rabbit Grimace Scale) assessed by video recordings in rabbits subjected to orthopedic surgery. The videos of 28 rabbits recorded as part of another study were used as follows: baseline (24 hours before surgery); pain (one hour after anesthesia recovery); analgesia (3 hours after analgesic rescue); and 24h (24 hours after anesthesic recovery). The RbtGS was blindly and randomly evaluated by three evaluators. The responsiveness was assessed through comparison of scores at the four time points using Friedman test followed by Dunn posthoc test. The influence of the observer at each time point was analyzed comparing the presence versus the absence of the observer using Wilcoxon test. Inter and intra-observer reliability was assessed through the intraclass correlation coefficient (ICC). The scale showed responsiveness because scores increased following the surgical procedure and decreased by 24 hours. The presence of an observer decreased RbtGS scores at the painful moment (presence: 0,75; 0-1,75, absence:1; 0-2) and increased at baseline (presence: 0,2; 0-2, absence:0; 0-2) and at 24h (presence: 0,33; 0-1,75, absence:0,2; 0-1,5). Intra- and inter-observer reliability ranged from moderate to very-good (ICC >0,49). The RbtGS showed responsiveness, repeatability and reproducibility in video assessment of rabbits subjected to orthopedic surgery. In the presence of a woman and familiar observer, the scores at the time when pain was greatest were underestimated and at times of little to no pain, the presence of an observer overestimated RbtGS scores, thereby suggesting that evaluations should be performed remotely. An important limitation of the RbtGS is the significant number of lost data.
In the last decades, facial pain scales have been used as tools to assess pain in several species. The goal of this study was to evaluate the responsiveness, the intra- and inter-observer reliability and the influence of a human observer on the scores of the RbtGS (Rabbit Grimace Scale) assessed by video recordings in rabbits subjected to orthopedic surgery. The videos of 28 rabbits recorded as part of another study were used as follows: baseline (24 hours before surgery); pain (one hour after anesthesia recovery); analgesia (3 hours after analgesic rescue); and 24h (24 hours after anesthesic recovery). The RbtGS was blindly and randomly evaluated by three evaluators. The responsiveness was assessed through comparison of scores at the four time points using Friedman test followed by Dunn posthoc test. The influence of the observer at each time point was analyzed comparing the presence versus the absence of the observer using Wilcoxon test. Inter and intra-observer reliability was assessed through the intraclass correlation coefficient (ICC). The scale showed responsiveness because scores increased following the surgical procedure and decreased by 24 hours. The presence of an observer decreased RbtGS scores at the painful moment (presence: 0,75; 0-1,75, absence:1; 0-2) and increased at baseline (presence: 0,2; 0-2, absence:0; 0-2) and at 24h (presence: 0,33; 0-1,75, absence:0,2; 0-1,5). Intra- and inter-observer reliability ranged from moderate to very-good (ICC >0,49). The RbtGS showed responsiveness, repeatability and reproducibility in video assessment of rabbits subjected to orthopedic surgery. In the presence of a woman and familiar observer, the scores at the time when pain was greatest were underestimated and at times of little to no pain, the presence of an observer overestimated RbtGS scores, thereby suggesting that evaluations should be performed remotely. An important limitation of the RbtGS is the significant number of lost data.
Descrição
Palavras-chave
Animais de laboratório, Avaliação da dor, Dor, Expressão facial, Facial expressions, Laboratory animals, Pain