| Objective:To explore the correlation between the characteristics of the tongue picture of patients with lung cancer and the occurrence of bone marrow suppression after chemotherapy,and combine the clinical characteristics to establish a prediction model for the occurrence of bone marrow suppression after chemotherapy of lung cancer,in order to provide a basis for clinical prevention and treatment of bone marrow suppression after chemotherapy.Methods:According to the established inclusion and exclusion criteria,103 patients who were hospitalized in the China-Japan Friendship Hospital and Beijing Friendship Hospital affiliated to the Capital Medical University between May 2019 and March 2021 were collected and screened.Before the start of the first cycle of chemotherapy,the DS01-G tongue examination instrument was used to collect the tongue picture of the patient,and the tongue picture characteristics were interpreted by two Chinese physicians with intermediate or above professional titles.The tongue picture characteristics interpretation content includes tongue color,tongue shape,color of tongue coating and texture of tongue coating,and then collate the results of tongue features for follow-up analysis;at the same time collect clinical information including patient age,gender,BMI,tumor stage and treatment drugs for follow-up analysis as well.After the completion of the first cycle of chemotherapy,the patient's bone marrow suppression was evaluated according to the "WHO Classification Standard for Common Toxic and Side Effects of Anticancer Drugs".The chi-square test was used to analyze the distribution of tongue features in people with and without bone marrow suppression.The univariate and multivariate logistic regression were used to study the correlation between tongue features and clinical factors and bone marrow suppression to screen the independent influencing factors of bone marrow suppression.Subsequently,the tongue characteristics and clinical factors were combined,and multivariate logistic regression was used to establish a prediction model of bone marrow suppression after the first cycle of lung cancer chemotherapy.The area under the ROC curve(AUC value)is used to discriminate the distinguishing ability of the model,the Hosmer-Lemeshow goodness of fit test and the calibration curve are used to evaluate the accuracy of the model,and finally the nomogram is used to visualize the model.Results:1.Among the tongue features of 103 patients with lung cancer,pale red tongue(40.8%)was the main tongue color distribution,and pale tongue(28.2%)was the main pathological tongue color,followed by red tongue(22.3%)and cyanoze tongue(8.7%);the main tongue shape distribution is normal tongue shape(43.7%),pathological tongue shape is mainly fat tongue(29.1%),followed by cracked tongue(19.4%)and tooth-marked tongue(13.6%),and finally the thin tongue(5.8%);the color distribution of the coated tongue is mainly white(49.5%)and yellow(50.5%).No patients with gray and black coated tongue were found in the current study;the tongue coating texture is mainly composed of thick greasy(60.2%),followed by thin moss(34.0%),and finally exfoliated tongue texture(5.8%)2.After the end of one cycle of chemotherapy,the overall incidence of bone marrow suppression was 39.8%,among which the incidence of grade ?,grade ?,and grade ? bone marrow suppression were 21.4%,11.7%,and 6.7%respectively,of which grade ? bone marrow suppression the incidence rate was 0.3.In the correlation analysis between tongue features and bone marrow suppression,the chi-square test results showed:fat tongue(x2=7.204,P=0.007)and tooth-marked tongue(x2=0.600,P=0.048)),which suggests that there is a difference in distribution between people with bone marrow suppression and those without bone marrow suppression;the results of univariate logistic regression show that fat tongue:OR=3.261,95%CI(1.366?8.052),P=0.009 and tooth-marked tongue:OR=3.206,95%CI(1.018?10.210),P=0.048,which is a risk factor for bone marrow suppression occurrence.Multivariate logistic regression analysis showed that fat tongue:OR=2.787,95%CI(2.319?3.349),P=0.027 is an independent risk factor for bone marrow suppression.4 In the correlation analysis between clinical features and bone marrow suppression,the results of univariate logistic regression showed that KPS score(>70 points):OR=0.649,95%CI(0.544?0.773),P<0.001 and highly toxic chemotherapeutics(>1 species):OR=3.859,95%CI(1.049?14.206),P=0.042 are the protective factors and risk factors for myelosuppression,respectively;the results of multivariate logistic regression are consistent with the results of univariate logistic regression:KPS score(>70 points):OR=0.649,95%CI(0.544?0.773),P<0.001;highly toxic chemotherapeutics(>1 species):OR=5.114,95%CI(1.139?22.965),P=0.033.5.Based on the tongue characteristics and clinical characteristics screened by multivariate logistic regression,a prediction model of bone marrow suppression after the first cycle of chemotherapy was established.The area under the ROC curve(AUC value)is 0.807.When the Youden index is at most-0.123,the model sensitivity is 73.2%,the specificity is 79%,suggesting that the model has better distinguishing ability.The results of the Hosmer-Lemeshow goodness-of-fit test showed:x2=8.398,P=0.396,indicating that there is no significant difference between the predicted incidence of bone marrow suppression after the first cycle and the actual incidence based on the model,indicating the goodness of the model better.At the same time,the results of the calibration curve showed a good agreement between the predicted incidence and the actual incidence,suggesting that the model has a better calibration.Finally,in order to facilitate the clinical operation,the nomogram is used to visualize the prediction model.Conclusion:Fat tongue and tooth-marked tongue are related to the occurrence of bone marrow suppression after the first cycle of chemotherapy in patients with lung cancer.Fat tongue is an independent risk factor for bone marrow suppression.In clinical features,KPS score greater than 70 is an independent protective factor for bone marrow suppression occurrence.The application of more than one type of highly toxic chemotherapeutics is an independent risk factor for bone marrow suppression occurrence;the bone marrow suppression prediction model established by combining the characteristics of tongue and clinical features includes three variables:fat tongue,KPS score,and highly toxic chemotherapy drugs.It indicates that its prediction performance and accuracy are better,and the nomogram established based on the model can visualize the model,which is convenient for clinical use. |
PDF Full Text Request