Orientación diagnóstica de la trombosis venosa profunda en la atención primaria. Estudio de coste-efectividad

Content Provider	Semantic Scholar
Author	Camps, F. Fondevila
Copyright Year	2015
Abstract	Introduccion: El diagnostico de la trombosis venosa profunda (TVP) es complejo debido a la inespecificidad de los signos y sintomas. Existen diferentes algoritmos para abordar el diagnostico de la TVP que incluyen el uso de escalas de probabilidad clinica, la determinacion del dimero-D y las pruebas de imagen. Sin embargo, actualmente la Atencion Primaria (AP) no utiliza estos algoritmos, situacion que obliga a derivar todos los pacientes con sospecha de TVP al hospital para descartar o confirmar el diagnostico. Objetivos: Determinar el algoritmo diagnostico mas coste-efectivo en los pacientes con sospecha de un primer episodio de TVP en la AP. Tambien se desea estimar el efecto de la aplicacion de algoritmos diagnosticos de TVP en el numero de derivaciones realizadas desde la AP a las urgencias especializadas y determinar las variables que se asocian a un diagnostico de confirmacion de TVP. Material y Metodos: Estudio observacional transversal de pacientes atendidos en urgencias hospitalarias, derivados desde la AP por sospecha de un primer episodio de TVP. Se analizaron tres posibles ramas diagnosticas: la rama habitual en la que se deriva sistematicamente a todos los pacientes con sospecha de TVP al hospital; el algoritmo de Wells basado en la escala de probabilidad clinica de Wells y la determinacion selectiva del dimero-D y/o ecografia doppler segun precise; la rama de Oudega, que analizaba la escala de Oudega que ya incluye el dimero-D como variable. En todos los pacientes se recogieron las variables de las escalas de probabilidad clinica (Wells y Oudega), se determino el valor del dimero-D (portatil y el del laboratorio hospitalario) y se realizo una ecografia doppler completa de extremidades inferiores. Se calcularon los costes directos imputados a cada una de las tres ramas diagnosticas analizadas. Se realizo la estadistica descriptiva y analitica, univariante y multivariante, y se determino el coste-efectividad de cada una de las tres ramas diagnosticas. Resultados: Se reclutaron 138 pacientes. De ellos, 22 fueron excluidos por no disponer de un informe de derivacion, por presentar clinica de mas de 30 dias, por estar anticoagulados o por presentar antecedentes de TVP previa. Se incluyeron finalmente 116 pacientes. El 61% eran mujeres y la edad media fue de 71 anos (DE:13,2). De los 116 pacientes incluidos, se confirmo el diagnostico de TVP en 22 casos (18,9%) mediante ecografia doppler completa. La presencia de dimero-D elevado se asocio con el diagnostico de TVP. El antecedente de enfermedad del aparato locomotor y la presencia de un probable diagnostico alternativo segun el clinico se relacionaron inversamente con el diagnostico de TVP. Las dos escalas de probabilidad clinica presentaron una sensibilidad del 100% y una especificidad igual o mayor al 40%. Utilizando la rama diagnostica habitual se derivan el 100% de los pacientes, mientras que con el algoritmo de Wells se hubiesen derivado un 64,7% y con la escala de Oudega un 67,2%. La utilizacion del algoritmo de Wells o de la escala de Oudega supondria una disminucion de costes de entre 9.998,92€ y 11.299,33€ en los pacientes estudiados. Conclusiones: La utilizacion de cualquiera de las dos escalas de probabilidad clinica y el dimero-D es mas coste-efectiva que la opcion diagnostica actual, consistente en derivar a todos los pacientes con sospecha clinica de TVP a urgencias del hospital. La opcion de utilizar la escala de Wells y su algoritmo diagnostico es mas coste-efectiva que el procedimiento diagnostico de Oudega. Los tres algoritmos presentan una elevada sensibilidad diagnostica. Con la aplicacion de algoritmos diagnosticos, se podria descartar con seguridad el diagnostico de TVP en uno de cada tres pacientes con sospecha de TVP atendidos en AP por esta patologia. Introduction: Diagnosis of deep vein thrombosis (DVT) is difficult due to the unspecific signs and symptoms. There are different algorithms to handle DVT diagnosis which include the use of clinical decision rules, the D-dimer test and imaging techniques. However, at present, primary care physicians do not use these algorithms and all patients have to be referred to the hospital to confirm or to rule out DVT diagnostic. Objectives: To determine the most cost-effective diagnostic algorithm in patients with symptoms compatible with a first DVT episode at primary care setting. To estimate the effect of the application of the DVT diagnostic algorithms on the number of referrals from primary care setting to hospital emergency rooms and to determine the items related to DVT diagnostic. Material and Methods: Observational cross-sectional study of patients visited at hospital emergency department referred from primary care with symptoms compatible with a first DVT episode. Three diagnostic options were defined: the present option in which all patients with symptoms compatible with DVT are systematically referred to the hospital; Wells’ algorithm which is based on his clinical prediction rule and the D-dimer test and/or Doppler ultrasound as needed; Oudega’s option, using his scale which already includes the D-dimer as a scale item. In all patients, the clinical decision rules items (Wells and Oudega) were recorded, the D-dimer test (portable and hospital laboratory) and a lower limbs complete Doppler ultrasound were performed. Direct costs generated by the three diagnostic algorithms were calculated. A descriptive and analytic statistical analysis (univariate and multivariate) was made. The cost-effectiveness of each diagnostic option was calculated. Results: 138 candidates were recruited and 22 patients were excluded because they did not have the referral report, they had symptoms for longer than 30 days, they were receiving anticoagulant treatment or had previous DVT antecedent. Finally 116 patients were included, 61% were women and the mean age was 71 years (SD:13.2). Of the 116 patients included, DVT was confirmed in 22 patients (18.9%) by using complete Doppler ultrasound. The positive D-dimer was related to the DVT diagnostic. The presence of musculoskeletal disease and the suspicion of a likely alternative diagnosis, according to physicians’ clinical judgment, were inversely associated with the DVT diagnosis. Both clinical decision rules showed a sensitivity of 100% and a specificity of 40% or higher. Using the present diagnostic option 100% of patients were referred to hospital. However, with Wells’ algorithm 64.7% would have been referred and 67.2% with Oudega’s algorithm. The diagnostic cost could have been reduced by 9.998,92€ to 11.299,33€ according to Oudega’s and Wells’ algorithm in all patients included. Conclusions: The use of any of the two clinical decision rules and the D-dimer is more cost-effective than the present algorithm which is based on the referral to hospital of all patients with clinical suspicion of DVT. The option of using Wells scale and Wells’ algorithm is more cost-effective than Oudega’s diagnostic option. The three diagnostic algorithms analysed have a high diagnostic sensitivity. With the application of these diagnostic algorithms, one third of all referrals to hospital emergency rooms from primary care setting, could have been avoided.
Starting Page	1
Ending Page	1
Page Count	1
File Format	PDF HTM / HTML
Alternate Webpage(s)	https://ddd.uab.cat/pub/tesis/2015/hdl_10803_401102/emfc1de1.pdf
Language	English
Access Restriction	Open
Content Type	Text
Resource Type	Article