Búsqueda de caso y diagnóstico temprano en sujetos con sospecha de la EPOC

Abstract

Rationale: Chronic obstructive pulmonary disease (COPD) is commonly underdiagnosed. To date, there is no a systematic evaluation of case finding in subjects with risk factors without symptoms (subjects >40 years with risk factor, no symptoms) and early diagnose those with symptoms (risk factors and symptoms) of the disease.

Methods: We performed a systematic review (PubMed, CINAHL, EMBASE) to identify observational studies that reported prevalence of COPD in case finding and early diagnose and calculated a weighted prevalence for each one of the groups. We also searched for studies that determined the accuracy of questionnaires and hand-held expiratory flow meter PiKo-6®, using spirometry as a gold standard (post-bronchodilator FEV1/FVC <70%) for the diagnosis of COPD. Two of the authors independently selected studies that met the inclusion criteria and assessed methodological quality using the GRADE approach. The Number Needed to Screen (NNS: 1/prevalence) to diagnose a case of COPD with spirometry was calculated for these two groups and compared with the value obtained from the national prevalence. Questionnaires and PiKo-6® were for case finding and the positive and negative posttest probability were calculated based on the weighted prevalence.

Results: For case finding, eleven studies met the inclusion criteria and the weighted prevalence in this group was 22%, with a NNS with spirometry of 5, compared with a NNS of 11 obtained from the prevalence of COPD of 8,9% (95% IC 8.2-9.7% ) in Colombia. After the critical evaluation of several questionnaires and studies of PiKo-6®, we selected the instrument developed by Müllernováet al. and the study of Frith et al., respectively. The validated questionnaire had a positive posttest probability of 56% and negative posttest probability of 3%, with NNS of 2. For PiKo-6®, the corresponding values were 44%, 7% and 2, respectively. For early diagnosis, the weighted prevalence from 3 studies was 33.9%, with a NNS with spirometry of 3.

Conclusion: For case finding of COPD, the NNS with spirometry is substantially lower than the NNS in the general population. The use of a questionnaire or the performance of PiKo-6® for decreases the NNS to 2 in the case finding group. Making early diagnosis in subjects with risk factor and symptoms is also an appropriate strategy with NNS of 3 for spirometry.