[Predictors of posttreatment olfactory improvement in patients with postviral olfactory dysfunction].

To analyzed the results of olfactory function test in patients with post-viral olfactory dysfunction（PVOD）, and evaluated the prognostic factors, so as to provide a basis for clinical diagnosis and treatment. This study included patients who were diagnosed with PVOD at least one year ago in Beijing Anzhen Hospital and whose telephone interviews of subjective olfactory function were available. The general condition of the patients, the results Sniffin' Sticks olfactory test and the event-related potentials（ERPs） were analyzed in different improvement groups. This study retrospectively analyzed PVOD patients treated in the outpatient department of Beijing Anzhen Hospital. They were given olfactory training for 4 months. The Sniffin' Sticks test was performed on the patients before and after the treatment. The Sniffin' Sticks test and event-related potentials（ERPs） results were used to evaluate the prognostic factors. In this study, the olfactory improvement rate of 63 PVOD patients was 52.38%（33/63）. Compared to the non-improvement group, the course of disease in the group with improved subjective olfactory function was significantly shorter（<0.001）, the initial olfactory function was significantly better（<0.001）, and the olfactory threshold was much lower（<0.001）. The presence of olfactory event-related potentials and trigeminal ERPs（tERPs） were 52.38%（33/63） and 87.30%（55/63）, respectively. The presence of oERPs in the olfactory function improvement group was significantly higher than that in the non-improvement group（<0.05）, but there was no difference in the presence of tERPs（>0.05）. Latency of N1 and P2 waves in oERPs with improvement group（ON1L, OP2L） were longer than those in the non-improvement group（<0.05）, N1 and P2 wave amplitudes（ON1A, OP2A） had no difference（>0.05）. The N1 and P2 amplitudes and latency of tERPs showed no difference between the two groups. Multivariate Logistic regression analysis showed that threshold value before treatment（=21.376, 95%: 2.172-210.377, =0.009）; ON1L（=0.994, 95%: 0.988-0.999, =0.029） and course of disease（=0.607, 95% 0.405-0.920, =0.016） was significantly associated with olfactory prognosis. The course of olfactory dysfunction, the severity of olfactory dysfunction, the threshold of olfactory function, and the latency of N1 wave of oERPs can be used to evaluate the prognosis of PVOD patients. However, age, olfactory discrimination, recognition ability, oERPs amplitude and tERPs wave value had less prognostic value.