The Analysis in Clinical Features and Correlated Factors of Fungal Rhino-sinusitis
|School||Dalian Medical University|
|Keywords||Fungal rhino-sinusitis Chronic rhino-sinusitis Diagnosis|
Objective： To summarize the statistically significant clinical features of fungalrhino-sinusitis （FRS） by comparing with the data of chronic rhino-sinusitis（CRS）,thenstudy on the correlated factors with statistics and built a predictive equation, helping therapid clinical diagnosis of FRS.Methods：Compared the102patients with FRS who were diagnosed by the ENTDepartment in the first Affiliated Hospital of Dalian Medical University from March2003to March2013and,102patients with CRS who were randomly sampled in thesame period and hospital. The clinical features and imaging features of FRS wereanalyzed by paired Chi-square test, and the correlated factors were investigated by usingthe multiple factor Logistic model regression analysis, then a predictive equation of theincidence of FRS was built. Finally, test the equation by the clinical data of204patientsof the study.Results：（1） Statistical analysis was performed on102cases of FRS and102cases ofCRS. The clinical characteristics of FRS were: Female, over40years of age, and courseof the disease<1year, with the symptoms of facial pain and haem-nasal, unilateral andunisinus morbidity in imaging, dense shadow calcified plaque and osteoproliferation inCT scan. While the clinical features of CRS were: Male, a long course of disease, withthe symptoms of nasal obstruction and hyposmia, nasal polyps in nasal cavity inspecialized medical. The following are the common performances both in FRS and CRSpatients: symptoms of discharging stuff, headache, cacosmia and signs of inferiorturbinate hypertrophy and purulent secretions in nasal passages. When the expoundedsigns and symptoms are found, rigid clinical examinations are needed.（2） Be analyzed by means of the Logistic model regression, seven correlated factorsin twenty-six were statistically significant: age, symptoms of haem-nasal, cacosmia,signs of inferior turbinate hypertrophy, unisinus lesions in imaging, calcified plaque andosteoproliferation in CT scan. The Logistic regression predictive equation of FRS was: y=10.968+2.294X1+1.890X2+2.595X3+2.074X4+3.286X5+4.373X6+4.010X7,P=exp（+y）/[1+exp（y）], in which the concomitant variables were age（X1）,haem-nasal（X2）, cacosmia（X3）, inferior turbinate（X4）, osteoproliferation（X5）, calcifiedplaque（X6）, unisinus lesion（X7）, respectively. The P value represents the probability ofsuffering from FRS.（3） The Logistic model regression equation was examined by the data of204patients involved in the study. Then the results were: the proporation of patients whogot P>0.50in FRS group occupied94.1%, those who got P>0.90accounted for85.3%.While in CRS group, those who got P>0.50was only8.8%.Conclusions：（1） The two diseases FRS and CRS can be identified by clinical featuresrespectively although their clinical presentations are exactly similar.（2） The multiple factor Logistic regression analysis could be used to build predictiveequation of disease after the clinical data of large sample were analyzed statistically inclinical research.（3） The earlier period and accurate diagnosis of FRS could be guided with the helpof predictive equation above.