Factors Influencing the Prognosis of Olfactory Dysfunction After Upper Sensation

All patients underwent nasal endoscopy and magnetic resonance imaging (MRI) of the olfactory pathway.

At least 1 week after the symptoms such as nasal congestion and runny nose subsided, olfactory disorders did not improve. Nasal sinusitis and other lesions were excluded by physical and endoscopic examinations and imaging (sinus computed tomography and MRI), and the disorder defined as post-infectious olfactory dysfunction. All participants had a clear history of upper respiratory tract infection; other causes of olfactory disorders were excluded.

All patients underwent a subjective olfactory test at the first visit and after the last treatment. The olfactory detection threshold and recognition threshold were tested using a T\&T olfactory meter (Japan's First Drug Industry Co., Ltd.)，which is commonly used in patients suffering from smelling disorders in asian, and correlate well with The University of Pennsylvania Smell Identification Test (UPSIT)[13]. The test was carried out according to the instructions on the olfactory meter, and the average recognition threshold of five olfactory elements was calculated to determine the degree of olfactory loss. The test results revealed normal (<1.0), mildly decreased (1.1 ≤ 2.5), moderately decreased (2.6 ≤ 4.0), severely decreased (4.1 ≤ 5.5), and complete lack of the sense of smell (> 5.5). In order to reduce the learning effects of repeated testing, all patients were order to close their eyes during the test.

The patients were treated with three protocols: (1) oral methylprednisolone, 40 mg for 3 days, 20 mg for 3 days, and 8 mg for 6 days; (2) oral methylprednisolone, 20 mg for 3 days, 10 mg for 3 days, and 4 mg for 6 days; and (3) budesonide atomization suspension (AstraZeneca Trading Co., Ltd.) inhaled through the nose atomized with an air compression atomizer for 4 weeks, 2 mg for the first 2 weeks and reduced to 1 mg for the last 2 weeks. At the same time, all patients received olfactory training for 12 weeks as proposed by Hummel et al[3].

A repeat T\&T subjective olfactory test will be performed after the treatment. Patients with normal sense of smell will be defined as cured; patients with average T\&T recognition threshold decreased by 1 point will be defined as improved; patients with average T\&T recognition threshold decreased by 2 points or more, which did not reach the normal range, will be defined as significantly improved; and patients with average T\&T recognition threshold decreased by 1 point, with or without falling, will be defined as invalid.

SAS 9.4 software was used for statistical analysis. The treatment plan was divided into three groups, and the dependent variable was a disordered three categorical variable. The description of the numerical variables of the normal distribution adopts the mean ± standard deviation, and the analysis of variance is used for statistical analysis; the description of the numerical variables of the non-normal distribution adopts the median (interquartile range), and the rank sum test of multi-sample comparison with a completely random design is adopted conduct statistical analysis. Categorical variables are described by R*C contingency table, and the chi-square test and Fisher exact probability method are used for statistical analysis. The alpha level of the statistical test is set to 0.05.