A composite of blood-based biomarkers to distinguish major depressive disorder and bipolar disorder in adolescents and adults

A composite of blood-based biomarkers to distinguish major depressive disorder and bipolar disorder in adolescents and adults | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article A composite of blood-based biomarkers to distinguish major depressive disorder and bipolar disorder in adolescents and adults Jieping Huang, Xuejiao Hou, Moyan Li, Yingshuang Xue, Jiangfei An, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3058571/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 16 Oct, 2023 Read the published version in BMC Psychiatry → Version 1 posted 10 You are reading this latest preprint version Abstract Objective The purpose of the present study was to construct a diagnostic model to distinguish major depressive disorder (MDD) and bipolar disorder (BD) using potential commonly tested blood biomarkers. Methods Information of 721 inpatients with an ICD-10 diagnosis of major depressive disorder or bipolar disorder were collected from the electronic medical record system. Variables in the nomogram were selected by best subset selection method after a prior univariable screening, and then constructed using logistic regression with inclusion of the psychotropic medication use. The discrimination, calibration and internal validation of the nomogram were evaluated by the receiver operating characteristic curve, the calibration curve, cross validation and subset validation method. Results The nomogram consisted of five variables, including age, eosinophil count, plasma concentrations of prolactin, total cholesterol, and low-density lipoprotein cholesterol. The model could discriminate between MDD and BD with an AUC of 0.858, with a sensitivity of 0.716 and a specificity of 0.890. Conclusion The comprehensive nomogram constructed by the present study can be convenient to distinguish MDD and BD since the incorporating variables were common indicators in clinical practice. It could help avoid misdiagnoses and improve prognosis of the patients. Major depressive disorder Bipolar disorder Diagnosis Nomogram Validation Figures Figure 1 Figure 2 Figure 3 Figure 4 Background Both major depressive disorder (MDD) and bipolar disorder (BD) are common chronic psychiatric disorders, which usually cause health damage or even severe disability to patients [ 1 ]. Accurate diagnosis of MDD and BD, which primarily depends on clinical manifestations, is the basis of individualized treatment and improvement of prognosis. However, the clinical features of MDD and BD often overlap, since the onset of BD is usually dominated by a depressive episode [ 2 ], and thus causes troubles to clinicians in diagnosis. Therefore, reducing misdiagnosis is critical to avoid delays of proper therapy or poorer outcomes. Although BD is a multifactorial disorder with several subtypes, such as bipolar disorder I (BD I) and bipolar disorder II (BD II), some large genome-wide association studies have found that no significant locus identified for BD overlapped with those identified for depression, while all BD subtypes have common variant heritability [ 3 ]. This provides a theoretical basis for the potential existence of biomarkers to distinguish between MDD and BP. Some studies have shown that indicators of individual systems, such as the blood system [ 4 ], the immune system [ 5 ], or multisystem biomarkers [ 6 , 7 ] may help improve diagnostic accuracy between MDD and certain types of BD such as BD I and BD II, or discriminate different phases of BD. However, given the complexity of etiology of mental diseases, biomarkers of a single system are unlikely to provide useful diagnostic effect size for MDD and BD, while the cost of detection of genes or protein profiles limits their use in clinical practice. Biomarkers from multiple routinely measured examinations can indicate functions or status of different systems and can also be easily obtained in clinical practice. We hypothesize that a panel of biomarkers combining routinely measured indicators might help differentiate between MDD and BP. In accordance, the aim of the study was to construct a prediction model to distinguish MDD and BD and evaluate its performance. Methods Study Population The study population consisted patients admitted to the Fifth Affiliated Hospital of Sun Yat-sen University from January 2019 to December 2021. In case of repeated hospitalizations, only the first admission was included. Information on qualified cases was extracted from the electronic medical record system after an ethical review by the Ethic Committee of the hospital. As a retrospective clinical study, the requirement for informed consent was exempted and identifiable personal information was removed to protect patient privacy. Inclusion and Exclusion Criteria The study included patients with an ICD-10 diagnosis of bipolar disorders (ICD-code: F31.) or depressive disorders (ICD-code: F32. & F33.). Although the ICD-codes include many subcategories that generally depend on the present clinical manifestation, however, we just used a simple diagnosis of large categories since some patients had phase changes during one hospitalization cycle. Exclusion criteria were pregnancy and severe diseases that might affect energy metabolism and immune responses, including chronic infectious diseases, inflammatory diseases, diabetes, any kind of malignant tumors or cancers, and neurodegenerative diseases. Data Collection We collected epidemiological data from the electronic medical record system, including age, gender, duration of the diseases, and marital status. And we also recorded the use of several psychotropic drugs, including antipsychotics, antidepressants, mood stabilizers, and benzodiazepines, in the month preceding the study entry, which we marked as present psychotropic medication use. In addition, results of routine blood tests were also collected. The selected blood tests were performed in the morning on the second hospitalization day, from a forearm vein after at least 10 hours of fasting. Predictors Selection Predictor selection was primarily performed using traditional statistical method and machine learning approaches. Firstly, an initial variable screening was performed using univariable analysis [ 8 ], and only covariates with a p -value of less than 0.01 were chosen for subsequent analysis. Secondly, predictors were further selected from the above variables using the best subset selection method, via the leaps package (Version 3.1) with complete cases [ 9 ]. During the process, we set the maximum size of the subset to eight, which was also the default number in the function. During repeated iterations, information criterions including Mallows’ Cp (CP), and Bayes Information Criteria (BIC) of different subset sizes were demonstrated in plots, all of which help determining the best subset [ 10 ]. Thirdly, decision curve analysis (DCA) was used to choose the final model when different information criterion directed to different best subsets [ 11 ]. Development of Prediction Model The prediction model was developed using logistic regression, and variables were excluded if their coefficients became insignificant after adjusting for the psychotropic medication use. Since variables were selected without considering observations with missing data, we used multiple imputation by chained equations (MICE) to avoid bias or inefficient estimates of parameters [ 12 ]. All results of the blood tests, in addition to age and gender, as well as the dichotomous outcome variable were included in the imputation. With the assumption that data were missing at random (MAR), the predictive mean matching (PMM) method was used to impute the missing variables using the mice package (version 3.15.0) in R. Prediction Model Presentation and Examination The prediction model was presented in the form of a nomogram and its accuracy was tested using the receiver operating characteristics (ROC) analysis [ 6 ], after deleting cases with missing data of the selected variables in the original study population. The area under the curve (AUC) with a higher value indicated higher accuracy. The model would be considered superior to a random ordering model if AUC > 0.5, while AUC value ranging from 0.5–0.7 indicate mild performance, 0.7–0.9 indicate moderate performance. In addition, sensitivity, specificity, and the likelihood ratios [ 13 ], including the positive likelihood ratio (LR [+]) and the negative likelihood ratio (LR [-]) were calculated to further test the accuracy of the model. After verifying the multivariable logistic model, we built a corresponding nomogram to provide a quantitative tool to help distinguish patients with MDD or BD. Calibration curves were plotted to assess the calibration of the nomogram [ 14 ], with the final regression model subjected to bootstrapping validation (1,000 bootstrap resamples), via the rms package (Version 6.3-0). Moreover, internal validation was performed using 10-fold cross-validation repeating 10 times [ 8 ], via caret package (Version 6.0–93). Moreover, different subsets were used to further validate the model, including the drug naïve subset, and the different age subgroups including 14–29 years, 30–44 years, and 45 + years. Statistical Analysis Data for continuous variables are presented as mean and standard deviation (SD), skewed data as median (25th and 75th percentiles), and categorical variables as absolute numbers and percentages. Clinical characteristics were compared using the Wilcoxon rank sum test for continuous variables and Pearson’s Chi-squared test, Fisher’s exact test when required, for categorical variables. All statistical analyses were conducted using the freely available statistical software R (version 4.2.0). The reported statistical significance levels were all two-sided, with an alpha value set at 0.05. Results Epidemiological Characteristics In total, 721 patients were included in the current study, 234 in the MDD group and 487 in the BD group. Characteristics of the study population are given in Table 1 . There were no statistically significant differences between patients with MDD and BD in gender, duration of illness, and family history of mental disorders, while patients of the two groups had different features in age, marital status, and use rate of antipsychotics, antidepressants, and mood stabilizers ( p < 0.01). Table 1 Epidemiological characteristics of patients with MDD or BP Variables Overall, N = 721 MDD, N = 234 BP, N = 487 p Age, years 21 (16, 37) 47 (27, 57) 17 (15, 24) < 0.001 Gender, No. (%) 0.061 Female 537 (74%) 164 (70%) 373 (77%) Male 184 (26%) 70 (30%) 114 (23%) Duration, years 2 (1, 4) 2 (0.5, 5) 2 (1, 4) 0.2 Family history, No. (%) 120 (17%) 38 (16%) 82 (17%) 0.8 Marital status, No. (%) < 0.001 Divorced 20 (2.8%) 8 (3.4%) 12 (2.5%) Married 223 (31%) 150 (64%) 73 (15%) Single 474 (66%) 72 (31%) 402 (83%) Widowed 4 (0.6%) 4 (1.7%) 0 (0.0%) Antipsychotics use, No. (%) 332 (46%) 68 (29%) 264 (54%) < 0.001 Antidepressants use, No. (%) 278 (39%) 124 (53%) 154 (32%) < 0.001 Mood stabilizers use, No. (%) 277 (38%) 19 (8.1%) 258 (53%) < 0.001 Benzodiazepines use, No. (%) 284 (39%) 101 (43%) 183 (38%) 0.2 Data are presented as median (IQR) or a number (percentage). P value is derived from univariable analyses using Wilcoxon rank sum test or Pearson's Chi-squared test (Fisher's exact test when needed). The psychotropic medication uses were limited to a month preceding study entry. Variable Selection Concerning results of blood tests, 226 (31.34%) participants had different degrees of data missing, most of which were concentrated on the examinations of the inflammatory and immune response (Supplement Figure). We did not delete incomplete observations directly but used complete cases for primary variable selection and then imputed incomplete cases for further validation. With preliminary data screening using univariable analysis, 22 potential biomarkers including age with a p -value < 0.01 were selected for best subset selection (Supplement Table). As demonstrated (Fig. 1 ), the subset with eight variables showed the smallest CP (Fig. 1 A). However, the regression coefficient for platelet-to-lymphocyte ratio (P.L) was not significant (p ≅ 0.09), so we decided to delete P.L from the model after verifying that its exclusion did not make a significant statistic difference using the ANOVA function in R. The subset with five variables showed the smallest BIC (Fig. 1 B). According to different information criterions, CP or BIC, a model with 7 or 5 specific variables could arguably be the best subset. Then how could we decide which model to choose? Here, we used DCA to compare the net benefit from the two diagnosis models, and it was shown that the latter one had moderately greater clinical benefits in general (Fig. 1 C), which consisted of age (unit: years), eosinophil count (Eos, unit: 10 9 /L), plasma concentrations of thyroid-stimulating hormone (TSH, unit: uIU/mL), follicle-stimulating hormone (FSH, unit: mIU/mL), prolactin (PRL, unit: ng/mL), total cholesterol (TC, unit: mmol/L), and low-density lipoprotein cholesterol (LDL, unit: mmol/L). Since complete case analysis may introduce bias, we used imputed datasets for consistency checks. If the conclusions drawn from the observatory cases or imputed cases were consistent, we could be confident that the conclusions were reliable. In addition, we briefly examined whether the present psychotropic medication use, including antipsychotics, antidepressants, mood stabilizers, and benzodiazepines, would influence the selected variables. It was shown that complete cases and imputed datasets provided similar results from multivariable logistic regression analysis (Table 2 ). After adjusting for present use of some psychotropic medications, the coefficients (β) and odd ratios (or exp(β)) of TSH and FSH became insignificant, which indicated these two indices were notably affected by the medications. Thus, we deleted TSH and FSH from the final model. Table 2 Multivariable regression for diagnosis between MDD and BP in patients Intercept and variables Model of complete cases Model of multiple imputations β Odds Ratio (95% CI) P β Odds Ratio (95% CI) P Intercept 2.539 < 0.001 2.587 < 0.001 EOS 2.358 10.6 (1.69 to 75.17) 0.015 2.392 10.9 (1.66 to 72.23) 0.013 TSH 0.102 1.10 (0.99 to 1.26) 0.104 0.104 1.11 (0.98 to 1.25) 0.098 FSH -0.014 0.99 (0.97 to 1.00) 0.068 -0.013 0.99 (0.97 to 1.00) 0.091 PRL 0.012 1.01 (1.00 to 1.02) 0.006 0.012 1.01 (1.00 to 1.02) 0.004 LDL 0.873 2.39 (1.29 to 4.54) 0.006 0.882 2.42 (1.30 to 4.49) 0.005 TC -0.698 0.50 (0.30 to 0.82) 0.007 -0.711 0.49 (0.30 to 0.81) 0.005 age -0.067 0.93 (0.92 to 0.95) < 0.001 -0.072 0.93 (0.91 to 0.95) < 0.001 β is the regression coefficient. Models are presented after adjustment of use of antipsychotics, antidepressants, mood stabilizers, and benzodiazepines. Presentation of Prediction Model Multivariable logistic regression analysis identified that age, LDL, TC, Eos, and PRL are independent predictors. The model incorporating these independent predictors was developed and presented as a nomogram (Fig. 2 ). Evaluation of Model Performance Model performance includes two characteristics, discrimination and calibration. Discrimination refers to the ability to distinguish between the two outcomes and can be assessed by concordance statistic (c-statistic). In logistic regression analysis, the c-statistic is equal to the area under the ROC curve (AUC). For the above nomogram, the c-statistic was 0.858, indicating good discrimination (Fig. 3 A). Moreover, a cutoff value of 0.66 showed a sensitivity of 0.716 and a specificity of 0.890. These values were calculated using observatory data containing 700 cases without missing values of selected variables. Moreover, LR [+] and LR [-] were 6.51 and 0.32, suggesting moderate shifts in probability of a correct diagnosis using the model. Calibration is used to evaluate the goodness of fit of the prediction model. The calibration plot here indicated that predicted probabilities approximately matched actual probabilities (Fig. 3 B). For this model, the Hosmer-Lemeshow test p -value was 0.705, indicating no evidence of poor model fit, in other words, the model seemed to fit well. Validation of Prediction Model Ten-fold cross validation was used for internal validation. The process of cross was repeated ten times (100 samples in total), and the average c-statistic, or AUC, across the samples was 0.853 (range from 0.850 to 0.856) (Fig. 4 A). This was close to but slightly lower than the overall model c-statistic of 0.858. The values showed the stability and reliability of predictions within the study population, and indicated good performance when the model is applied to new patients. Moreover, subset validation with ROC curve was used to further evaluate the effectiveness of the model. In the drug naïve subset, the AUC was 0.826, indicating good discrimination (Fig. 4 B). In different age subsets, the AUC ranged from 0.671 to 0.739, indicating mild to moderate discrimination (Fig. 4C1-C3). Discussion After many years of effort, researchers have not yet constructed a prediction model for discriminating between BD and MDD with clinical utility. In the present study, we preliminarily developed and validated a diagnostic nomogram, with a composite of biomarkers from routinely tested blood results, to distinguish MDD and BD. The model was constructed using the best subset selection method and then verified and adjusted using multiple imputations and the inclusion of psychotropic medication use. The final model consisted of five variables: age, LDL, TC, Eos, and PRL. The model could discriminate between MDD and BD with an AUC of 0.858, with a sensitivity of 0.716 and a specificity of 0.890. During the construction of the model, 47 features were reduced to 22 potential predictors at the first step by univariant analysis, then the best subset selection method was managed to select seven prominent markers. Of the 721 patients in the study, only 495 without missing data were used for the primary multivariable selections. Then 700 were used for adjustment and evaluation of the model after deleting cases with incomplete values of the prominent variables, which made the findings relatively more robust than constructing and validating the model using the same population. Moreover, repeated cross validations were subsequently used to verify the model when the training dataset and test dataset did not overlap, and subset validations were used to test the effectiveness of the model in drug naïve patients and patients of different age groups. The findings of the present study were somewhat consistent with previous studies. For example, age is one of the most profound distinguishing factors between MDD and BD, as it had been broadly accepted that the onset age of MDD is generally later than that of BD [ 15 , 16 ]. However, we wanted to see how the performance of the composite biomarkers would change if the effect of age was minimized. The study divided patients into three groups, 14–29 years, 30–44 years group, and 45 + years respectively. Within each group, age became insignificant different between MDD and BP patients (data not shown). Unsurprisingly, the model discrimination had varying degrees of deteriorations, and the AUC were 0.688, 0.671, and 0.739 respectively, indicating that the model still had mild to moderate diagnostic efficiency in patients of same age group. Moreover, eosinophil could also help discriminate the two disorders, which was consistent with previous studies. For example, it has been demonstrated that eosinophil counts were reduced in MDD patients [ 17 ], while the increased eosinophil function could be found in the late-stage of BD [ 18 ]. In addition, the inclusion of PRL in the model, one of the hormones secreted by pituitary gland, suggested that pituitary function might play a role in differentiating MDD and BD. However, previous studies on the pituitary gland mainly focused on the gland volume changes in mental disorders and their association with hyperactivity in the Hypothalamic-Pituitary-Adrenal axis [ 19 – 21 ]. Other hormones provided by pituitary gland besides adrenocorticotropic hormone could also have potential effects on mental disorders. In this study, TSH and FSH were tested statistically significant but excluded after preliminary adjustment for psychotropic medication use, which was in accordance with clinical consensus that endocrine is greatly influenced during the drug treatment for affective disorders [ 22 ]. Interestingly, PRL remained in the model after medication adjustment. However, these findings require further confirmation in drug-free patients. In addition, LDL and TC were also included in the final model. These findings did not contradict previous findings that abnormal lipid metabolism was more prevalent in MDD and BD patients compared to health controls [ 23 , 24 ]. However, few studies have compared the differences in lipid profile distribution between MDD and BD. Our study showed that BD patients had relatively higher LDL levels, while MDD patients had higher TC levels. Although these findings indicated different lipid profiles in MDD and BP patients, but both were consistent with the findings that patients with severe mental illnesses had increased risks for cardiovascular diseases [ 25 , 26 ]. Moreover, like endocrine functions, lipid metabolism is also seriously affected by some kinds of psychotropic drugs, especially antipsychotics and mood stabilizers, such as clozapine, olanzapine, and valproate [ 27 ], which can ultimately result in hyperlipidemia or even obesity. As it was demonstrated in Table 1 , the proportion of BD patients using antipsychotics and mood stabilizers was significantly higher than that of MDD patients, however, the coefficients of TC and LDL in the regression model remained significant after the adjustment of medication use in this study, indicating that the pharmacological effect was not the only reason for the differences in the lipid levels between the two groups. In other words, abnormal lipid metabolism may underlie the mental disorders. However, since the cholesterol level can be greatly influenced by living habits, such as diet and physical activities [ 28 ], the significance could not be applied to a population with different lifestyles. Emerging studies have confirmed the potential roles of inflammation or immune-based biomarkers as predictive biomarker panels to differentiate MDD and BD, usually including C-reactive protein (CRP), interleukins, and complement components [ 5 , 29 – 31 ]. However, the above related potential biomarkers were surprisingly excluded during model development, which was inconsistent with previous findings. For example, Chang et.al demonstrated that baseline C-reactive protein could serve as a discrimination biomarker for MDD and bipolar II disorder in drug naïve patients (cutoff value: 621.6 ng/mL; AUC value: 0.816), and patients with baseline C-reactive protein greater than 621.6 ng/mL had 28.2 higher odds of bipolar II disorder [ 32 ]. However, in our study, C-reactive protein level showed no statistical difference between MDD and BD and was excluded at the first step. The possible reason might be treatment effects as indicated by Chang’s work itself: the difference of C-reactive protein level would become narrower between MDD and bipolar II disorder after treatment. Another possible reason may be bias from concentrated missing values on inflammation and immune factors; although the multiple imputations had indicated that the missingness of the selected variables in the model was at random, it may not represent the same missing pattern of the potential predictors in question [ 33 ]. There were several limitations to this study. Firstly, behavior characteristics and psychological assessments failed to be included in the analysis process. Secondly, BD patients were not specifically classified into different clinical phases including (hypo)manic or depressive phase, or rapid cycling BD. Moreover, although the diagnostic biomarker panel constructed by this study was preliminarily unlikely to be affected by medications as cofounding factors but should be tested further in drug naïve patients. At last, the data were collected from one hospital, the generalizability of the prediction panel needs further testing. Conclusion Our study presents a nomogram that incorporates factors from commonly tested blood indicators that can conveniently help distinguishing MDD and BD, and thus reduce misdiagnosis. Declarations Authors’ contributions Jieping Huang designed the study, analyzed the data, and wrote the manuscript. Xuejiao Hou, Moyan Li, Yingshuang Xue collected the data. Jiangfei An, Shenglin Wen and Zi Wang reviewed the writing of the manuscript. Minfeng Cheng and Jihui Yue supervised the study. All authors contributed to and have approved the submitted manuscript. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Availability of Data and Materials The data of this study cannot be made publicly available for confidentiality reasons. Data are however available from the corresponding author upon reasonable request. Acknowledgement Not applicable. Competing interests The authors declare no financial or other conflict of interest. Consent for publication Not applicable. Ethical approval and consent to participate The research was conducted according to the International Conference on Harmonization – Good Clinical Practice Guidelines and the Declaration of Helsinki. The study procedures were reviewed and approved by the Medical Ethic Committee Review Board of the Fifth Affiliated Hospital of Sun Yat-sen University (Approval number: [2022] K44-1). The need for informed consent was waived by the Medical Ethic Committee Review Board of the Fifth Affiliated Hospital of Sun Yat-sen University. This was because the research used human materials or data with identifiable personal information, but the subject could no longer be found and the research project did not involve personal privacy or commercial interests, according to the Article 39 of the ‘Ethical Review Measures for Biomedical Research Involving Humans’ issued by the National Health and Family Planning Commission of the People’s Republic of China. References N. H. Kalin, "New Insights Into Major Depression and the Treatment of Bipolar Depression," (in eng), Am J Psychiatry, vol. 178, no. 12, pp. 1071-1074, Dec 2021, doi: 10.1176/appi.ajp.2021.21101042. I. Grande, M. Berk, B. Birmaher, and E. Vieta, "Bipolar disorder," (in eng), Lancet, vol. 387, no. 10027, pp. 1561-1572, Apr 09 2016, doi: 10.1016/S0140-6736(15)00241-X. E. A. Stahl et al. , "Genome-wide association study identifies 30 loci associated with bipolar disorder," (in eng), Nat Genet, vol. 51, no. 5, pp. 793-803, May 2019, doi: 10.1038/s41588-019-0397-8. L. Fusar-Poli et al. , "Neutrophil-to-Lymphocyte, Platelet-to-Lymphocyte and Monocyte-to-Lymphocyte Ratio in Bipolar Disorder," (in eng), Brain Sci, vol. 11, no. 1, Jan 06 2021, doi: 10.3390/brainsci11010058. B. I. Perry, R. Upthegrove, N. Kappelmann, P. B. Jones, S. Burgess, and G. M. Khandaker, "Associations of immunological proteins/traits with schizophrenia, major depression and bipolar disorder: A bi-directional two-sample mendelian randomization study," (in eng), Brain Behav Immun, vol. 97, pp. 176-185, Oct 2021, doi: 10.1016/j.bbi.2021.07.009. K. Munkholm, M. Vinberg, B. K. Pedersen, H. E. Poulsen, C. T. Ekstrøm, and L. V. Kessing, "A multisystem composite biomarker as a preliminary diagnostic test in bipolar disorder," (in eng), Acta Psychiatr Scand, vol. 139, no. 3, pp. 227-236, Mar 2019, doi: 10.1111/acps.12983. S. J. Rhee et al. , "Comparison of serum protein profiles between major depressive disorder and bipolar disorder," (in eng), BMC Psychiatry, vol. 20, no. 1, p. 145, Apr 03 2020, doi: 10.1186/s12888-020-02540-0. E. H. Au, A. Francis, A. Bernier-Jean, and A. Teixeira-Pinto, "Prediction modeling-part 1: regression modeling," (in eng), Kidney Int, vol. 97, no. 5, pp. 877-884, May 2020, doi: 10.1016/j.kint.2020.02.007. G. Heinze, C. Wallisch, and D. Dunkler, "Variable selection - A review and recommendations for the practicing statistician," (in eng), Biom J, vol. 60, no. 3, pp. 431-449, May 2018, doi: 10.1002/bimj.201700067. R. I. Kabacoff, R in Action , 2 ed. New York, NY: Manning Publications, 2015. M. Fitzgerald, B. R. Saville, and R. J. Lewis, "Decision curve analysis," (in eng), JAMA, vol. 313, no. 4, pp. 409-10, Jan 27 2015, doi: 10.1001/jama.2015.37. K. Blazek, A. van Zwieten, V. Saglimbene, and A. Teixeira-Pinto, "A practical guide to multiple imputation of missing data in nephrology," (in eng), Kidney Int, vol. 99, no. 1, pp. 68-74, Jan 2021, doi: 10.1016/j.kint.2020.07.035. M. Weatherall, "Information provided by diagnostic and screening tests: improving probabilities," (in eng), Postgrad Med J, vol. 94, no. 1110, pp. 230-235, Apr 2018, doi: 10.1136/postgradmedj-2017-135273. Y. Q. Huang et al. , "Development and Validation of a Radiomics Nomogram for Preoperative Prediction of Lymph Node Metastasis in Colorectal Cancer," (in eng), J Clin Oncol, vol. 34, no. 18, pp. 2157-64, Jun 20 2016, doi: 10.1200/JCO.2015.65.9128. E. Vieta et al. , "Early Intervention in Bipolar Disorder," (in eng), Am J Psychiatry, vol. 175, no. 5, pp. 411-426, May 01 2018, doi: 10.1176/appi.ajp.2017.17090972. L. T. Park and C. A. Zarate, "Depression in the Primary Care Setting," (in eng), N Engl J Med, vol. 380, no. 6, pp. 559-568, Feb 07 2019, doi: 10.1056/NEJMcp1712493. D. Singh et al. , "Changes in leukocytes and CRP in different stages of major depression," (in eng), J Neuroinflammation, vol. 19, no. 1, p. 74, Apr 04 2022, doi: 10.1186/s12974-022-02429-7. B. Panizzutti et al. , "Increased serum levels of eotaxin/CCL11 in late-stage patients with bipolar disorder: An accelerated aging biomarker?," (in eng), J Affect Disord, vol. 182, pp. 64-9, Aug 15 2015, doi: 10.1016/j.jad.2014.12.010. G. Delvecchio et al. , "Pituitary gland shrinkage in bipolar disorder: The role of gender," (in eng), Compr Psychiatry, vol. 82, pp. 95-99, Apr 2018, doi: 10.1016/j.comppsych.2018.01.014. G. Delvecchio, A. C. Altamura, J. C. Soares, and P. Brambilla, "Pituitary gland in Bipolar Disorder and Major Depression: Evidence from structural MRI studies: Special Section on "Translational and Neuroscience Studies in Affective Disorders". Section Editor, Maria Nobile MD, PhD. This Section of JAD focuses on the relevance of translational and neuroscience studies in providing a better understanding of the neural basis of affective disorders. The main aim is to briefly summarise relevant research findings in clinical neuroscience with particular regards to specific innovative topics in mood and anxiety disorders," (in eng), J Affect Disord, vol. 218, pp. 446-450, Aug 15 2017, doi: 10.1016/j.jad.2017.03.066. J. Keller et al. , "HPA axis in major depression: cortisol, clinical symptomatology and genetic variation predict cognition," (in eng), Mol Psychiatry, vol. 22, no. 4, pp. 527-536, Apr 2017, doi: 10.1038/mp.2016.120. J. F. Hayes, L. Marston, K. Walters, J. R. Geddes, M. King, and D. P. Osborn, "Adverse Renal, Endocrine, Hepatic, and Metabolic Events during Maintenance Mood Stabilizer Treatment for Bipolar Disorder: A Population-Based Cohort Study," (in eng), PLoS Med, vol. 13, no. 8, p. e1002058, Aug 2016, doi: 10.1371/journal.pmed.1002058. Y. G. Wei et al. , "Cholesterol and triglyceride levels in first-episode patients with major depressive disorder: A meta-analysis of case-control studies," J Affect Disord, vol. 266, pp. 465-472, Apr 1 2020, doi: 10.1016/j.jad.2020.01.114. N. Dalkner et al. , "Metabolic Syndrome Impairs Executive Function in Bipolar Disorder," (in eng), Front Neurosci, vol. 15, p. 717824, 2021, doi: 10.3389/fnins.2021.717824. B. I. Goldstein et al. , "Major Depressive Disorder and Bipolar Disorder Predispose Youth to Accelerated Atherosclerosis and Early Cardiovascular Disease: A Scientific Statement From the American Heart Association," (in eng), Circulation, vol. 132, no. 10, pp. 965-86, Sep 08 2015, doi: 10.1161/CIR.0000000000000229. R. E. Nielsen, J. Banner, and S. E. Jensen, "Cardiovascular disease in patients with severe mental illness," (in eng), Nat Rev Cardiol, vol. 18, no. 2, pp. 136-145, Feb 2021, doi: 10.1038/s41569-020-00463-7. T. Pillinger et al. , "Comparative effects of 18 antipsychotics on metabolic function in patients with schizophrenia, predictors of metabolic dysregulation, and association with psychopathology: a systematic review and network meta-analysis," (in eng), Lancet Psychiatry, vol. 7, no. 1, pp. 64-77, Jan 2020, doi: 10.1016/S2215-0366(19)30416-X. U. E. Bauer, P. A. Briss, R. A. Goodman, and B. A. Bowman, "Prevention of chronic disease in the 21st century: elimination of the leading preventable causes of premature death and disability in the USA," (in eng), Lancet, vol. 384, no. 9937, pp. 45-52, Jul 05 2014, doi: 10.1016/S0140-6736(14)60648-6. G. R. Fries, M. J. Zamzow, T. Andrews, O. Pink, G. Scaini, and J. Quevedo, "Accelerated aging in bipolar disorder: A comprehensive review of molecular findings and their clinical implications," (in eng), Neurosci Biobehav Rev, vol. 112, pp. 107-116, May 2020, doi: 10.1016/j.neubiorev.2020.01.035. F. Haenisch et al. , "Towards a blood-based diagnostic panel for bipolar disorder," (in eng), Brain Behav Immun, vol. 52, pp. 49-57, Feb 2016, doi: 10.1016/j.bbi.2015.10.001. H. Yu et al. , "Association of the plasma complement system with brain volume deficits in bipolar and major depressive disorders," (in eng), Psychol Med, pp. 1-11, Oct 26 2022, doi: 10.1017/S0033291722003282. H. H. Chang et al. , "C-reactive protein: A differential biomarker for major depressive disorder and bipolar II disorder," (in eng), World J Biol Psychiatry, vol. 18, no. 1, pp. 63-70, Feb 2017, doi: 10.3109/15622975.2016.1155746. I. R. White, P. Royston, and A. M. Wood, "Multiple imputation using chained equations: Issues and guidance for practice," (in eng), Stat Med, vol. 30, no. 4, pp. 377-99, Feb 20 2011, doi: 10.1002/sim.4067. Additional Declarations No competing interests reported. Supplementary Files supplementFigure.tif supplementfigurelegend.docx supplementtable.docx Cite Share Download PDF Status: Published Journal Publication published 16 Oct, 2023 Read the published version in BMC Psychiatry → Version 1 posted Editorial decision: Major revision 21 Aug, 2023 Reviews received at journal 15 Aug, 2023 Reviewers agreed at journal 14 Aug, 2023 Reviewers agreed at journal 13 Aug, 2023 Reviewers agreed at journal 11 Aug, 2023 Reviewers invited by journal 04 Jul, 2023 Editor assigned by journal 04 Jul, 2023 Editor invited by journal 22 Jun, 2023 Submission checks completed at journal 22 Jun, 2023 First submitted to journal 13 Jun, 2023 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About In Review Editorial Policies Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3058571","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":212106037,"identity":"759a0a62-dde1-446a-8d85-9a2ea8a322ec","order_by":0,"name":"Jieping Huang","email":"","orcid":"","institution":"The Fifth Affiliated Hospital of Sun Yat-sen University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Jieping","middleName":"","lastName":"Huang","suffix":""},{"id":212106038,"identity":"bdfee99d-7a08-4973-af57-1b5dbcfed96f","order_by":1,"name":"Xuejiao Hou","email":"","orcid":"","institution":"The Fifth Affiliated Hospital of Sun Yat-sen University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Xuejiao","middleName":"","lastName":"Hou","suffix":""},{"id":212106039,"identity":"bd2a5102-dbba-4d0d-800a-52049a08eb0f","order_by":2,"name":"Moyan Li","email":"","orcid":"","institution":"The Fifth Affiliated Hospital of Sun Yat-sen University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Moyan","middleName":"","lastName":"Li","suffix":""},{"id":212106040,"identity":"ddd402b4-52c3-4e74-bea0-8651e5aaef94","order_by":3,"name":"Yingshuang Xue","email":"","orcid":"","institution":"The Fifth Affiliated Hospital of Sun Yat-sen University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yingshuang","middleName":"","lastName":"Xue","suffix":""},{"id":212106041,"identity":"2642ff5a-075f-47e3-ac8f-7082c5f23a04","order_by":4,"name":"Jiangfei An","email":"","orcid":"","institution":"The Fifth Affiliated Hospital of Sun Yat-sen University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Jiangfei","middleName":"","lastName":"An","suffix":""},{"id":212106043,"identity":"82dd38db-08db-416d-816c-af4351404fb4","order_by":5,"name":"Shenglin Wen","email":"","orcid":"","institution":"The Fifth Affiliated Hospital of Sun Yat-sen University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Shenglin","middleName":"","lastName":"Wen","suffix":""},{"id":212106044,"identity":"d24d05e7-3f5b-4d77-8166-da3b5f5738b7","order_by":6,"name":"Zi Wang","email":"","orcid":"","institution":"Zhuhai Promotion Association of Mental Health","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Zi","middleName":"","lastName":"Wang","suffix":""},{"id":212106046,"identity":"65d01002-9719-4444-b2b6-776cf6b61afc","order_by":7,"name":"Minfeng Cheng","email":"","orcid":"","institution":"The Fifth Affiliated Hospital of Sun Yat-sen University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Minfeng","middleName":"","lastName":"Cheng","suffix":""},{"id":212106047,"identity":"25a42ef6-c7d8-4497-865c-1f9327c93a3d","order_by":8,"name":"Jihui Yue","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8klEQVRIiWNgGAWjYJACY4YKBgY2djg/gRgtZ4BamEnRwszYBiKJ1WJwI/lAceG8bfJ8zAzMn3n+HGbgZ88xYPi5A5+WtATjmdtuG7YxM7BJ87YdZpDseWPA2HsGn5YcA2PebbcZQVqYeRsOg0UgTsWpJf+DMe+c2/ZtMIfZE9aSw2DM23A7EaiFQZqHDWiLBAEtkmeeGRjzHLud3AZUJjm3LZ1H4syzgoO9eLTwHU9+ZsxTc9t2fnvz4Q9v/ljL8bcnb3zwE48WhQMMbAYQJmMDEw8DAw+IeQC3BgYG+QYG5gcwDuMPfEpHwSgYBaNgxAIASYZL66Xsk9AAAAAASUVORK5CYII=","orcid":"","institution":"The Fifth Affiliated Hospital of Sun Yat-sen University","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Jihui","middleName":"","lastName":"Yue","suffix":""}],"badges":[],"createdAt":"2023-06-13 13:59:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3058571/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3058571/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12888-023-05204-x","type":"published","date":"2023-10-16T15:01:18+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":39130545,"identity":"bf257a3c-328c-436f-8b90-bbbc0a624a3c","added_by":"auto","created_at":"2023-06-26 23:17:30","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":52061,"visible":true,"origin":"","legend":"\u003cp\u003e(A-B) Best models for each subset size based on Mallows’ Cp (CP) and Bayes Information Criteria (BIC). (C) Decision curve analysis for the model with 5 variables (sub.fit.5) and the model with 7 variables (sub.fit.7). \u003cem\u003eWBC, white blood cell count; PLT: platelet count; Lym, lymphocyte count; Eos, eosinophil count; P.L, platelet-to-lymphocyte ratio; N.L, neutrophil-to-lymphocyte ratio; E.L, eosinophil-to-lymphocyte ratio; FT3, free triiodothyronine; FT3.FT4, free triiodothyronine-to-free thyroxine ratio; TSH, thyroid stimulating hormone; FSH, follicle stimulating hormone; E2, estradiol; PRL, prolactin; ALT, alanine transaminase; UA, uric acid; LDL, low density lipoprotein cholesterol; TC, total cholesterol; IgA, immunoglobulin A; ALB, albumin.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"OnlineFigure1.png","url":"https://assets-eu.researchsquare.com/files/rs-3058571/v1/1d514d3427bebf6ef2e902a5.png"},{"id":39130546,"identity":"e0f7d4e1-9414-45b0-9c6e-f248e59ec7c4","added_by":"auto","created_at":"2023-06-26 23:17:30","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":42011,"visible":true,"origin":"","legend":"\u003cp\u003eThe nomogram developed in the observatory populations, incorporating age, total cholesterol (TC), and low-density lipoprotein cholesterol (LDL), eosinophil counts (Eos), and prolactin (PRL).\u003c/p\u003e","description":"","filename":"OnlineFigure2.png","url":"https://assets-eu.researchsquare.com/files/rs-3058571/v1/8c99a9802952d10b3202e7aa.png"},{"id":39130547,"identity":"3c5e989c-9000-4756-b2c9-8c942af54979","added_by":"auto","created_at":"2023-06-26 23:17:30","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":62649,"visible":true,"origin":"","legend":"\u003cp\u003e(A) Receiver operator characteristics (ROC) curve for the diagnostic model to distinguish patients with MDD or BD. For logistic regression models, c-statistic is equal to Area Under the ROC Curve (AUC). (B)Calibration curve. The x-axis represents the predicted probability and y-axis represents the actual probability of BD diagnosis. Perfect prediction would correspond to the 45° dashed line, the dotted line represents the observatory cases (n=700), the solid line is bias-corrected by bootstrapping (B = 1000 repetitions).\u003c/p\u003e","description":"","filename":"OnlineFigure3.png","url":"https://assets-eu.researchsquare.com/files/rs-3058571/v1/4ecff65ad7cc296010c05c50.png"},{"id":39130550,"identity":"bf8bddf1-493f-4d0a-9c08-05bf0c0dda7d","added_by":"auto","created_at":"2023-06-26 23:17:31","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":164215,"visible":true,"origin":"","legend":"\u003cp\u003e(A) Box plot showing the results of average AUC, or c- statistic, on the 100 cross-validation samples (10-fold cross validation repeated 10 times). (B, C1-C3) ROC curve distinguished patients with MDD or BD in different subgroups, including drug naïve patients (B) and different age groups (C1-C3), 14-29 year group, 30-44 year group, and 45+ year group respectively.\u003c/p\u003e","description":"","filename":"OnlineFigure4.png","url":"https://assets-eu.researchsquare.com/files/rs-3058571/v1/24503a5a8ba1f0f365b075b3.png"},{"id":45090927,"identity":"4c730936-5cc8-4c3c-bd13-a54ecf60e16f","added_by":"auto","created_at":"2023-10-23 15:07:34","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":625815,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3058571/v1/b7fa6b2f-7479-4b56-974c-e50fd932106a.pdf"},{"id":39131769,"identity":"77b576c4-5ba7-45bc-8a26-4d75c949e945","added_by":"auto","created_at":"2023-06-26 23:25:31","extension":"tif","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1047344,"visible":true,"origin":"","legend":"","description":"","filename":"supplementFigure.tif","url":"https://assets-eu.researchsquare.com/files/rs-3058571/v1/d72c533e6f8c14026b692ae5.tif"},{"id":39131768,"identity":"35abced2-a624-4b46-9ada-8f5de1f8458b","added_by":"auto","created_at":"2023-06-26 23:25:31","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":18900,"visible":true,"origin":"","legend":"","description":"","filename":"supplementfigurelegend.docx","url":"https://assets-eu.researchsquare.com/files/rs-3058571/v1/39ed9415ea69f473e6a0622c.docx"},{"id":39130549,"identity":"d7f7f385-c958-492f-b6e1-7bc7378dbe1e","added_by":"auto","created_at":"2023-06-26 23:17:31","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":25146,"visible":true,"origin":"","legend":"","description":"","filename":"supplementtable.docx","url":"https://assets-eu.researchsquare.com/files/rs-3058571/v1/ca51ee8ac568c5a833cb8670.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"A composite of blood-based biomarkers to distinguish major depressive disorder and bipolar disorder in adolescents and adults","fulltext":[{"header":"Background","content":"\u003cp\u003eBoth major depressive disorder (MDD) and bipolar disorder (BD) are common chronic psychiatric disorders, which usually cause health damage or even severe disability to patients [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Accurate diagnosis of MDD and BD, which primarily depends on clinical manifestations, is the basis of individualized treatment and improvement of prognosis. However, the clinical features of MDD and BD often overlap, since the onset of BD is usually dominated by a depressive episode [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e], and thus causes troubles to clinicians in diagnosis. Therefore, reducing misdiagnosis is critical to avoid delays of proper therapy or poorer outcomes.\u003c/p\u003e \u003cp\u003eAlthough BD is a multifactorial disorder with several subtypes, such as bipolar disorder I (BD I) and bipolar disorder II (BD II), some large genome-wide association studies have found that no significant locus identified for BD overlapped with those identified for depression, while all BD subtypes have common variant heritability [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. This provides a theoretical basis for the potential existence of biomarkers to distinguish between MDD and BP. Some studies have shown that indicators of individual systems, such as the blood system [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e], the immune system [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e], or multisystem biomarkers [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e] may help improve diagnostic accuracy between MDD and certain types of BD such as BD I and BD II, or discriminate different phases of BD. However, given the complexity of etiology of mental diseases, biomarkers of a single system are unlikely to provide useful diagnostic effect size for MDD and BD, while the cost of detection of genes or protein profiles limits their use in clinical practice.\u003c/p\u003e \u003cp\u003eBiomarkers from multiple routinely measured examinations can indicate functions or status of different systems and can also be easily obtained in clinical practice. We hypothesize that a panel of biomarkers combining routinely measured indicators might help differentiate between MDD and BP. In accordance, the aim of the study was to construct a prediction model to distinguish MDD and BD and evaluate its performance.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy Population\u003c/h2\u003e \u003cp\u003eThe study population consisted patients admitted to the Fifth Affiliated Hospital of Sun Yat-sen University from January 2019 to December 2021. In case of repeated hospitalizations, only the first admission was included. Information on qualified cases was extracted from the electronic medical record system after an ethical review by the Ethic Committee of the hospital. As a retrospective clinical study, the requirement for informed consent was exempted and identifiable personal information was removed to protect patient privacy.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eInclusion and Exclusion Criteria\u003c/h2\u003e \u003cp\u003eThe study included patients with an ICD-10 diagnosis of bipolar disorders (ICD-code: F31.) or depressive disorders (ICD-code: F32. \u0026amp; F33.). Although the ICD-codes include many subcategories that generally depend on the present clinical manifestation, however, we just used a simple diagnosis of large categories since some patients had phase changes during one hospitalization cycle. Exclusion criteria were pregnancy and severe diseases that might affect energy metabolism and immune responses, including chronic infectious diseases, inflammatory diseases, diabetes, any kind of malignant tumors or cancers, and neurodegenerative diseases.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eData Collection\u003c/h2\u003e \u003cp\u003eWe collected epidemiological data from the electronic medical record system, including age, gender, duration of the diseases, and marital status. And we also recorded the use of several psychotropic drugs, including antipsychotics, antidepressants, mood stabilizers, and benzodiazepines, in the month preceding the study entry, which we marked as present psychotropic medication use. In addition, results of routine blood tests were also collected. The selected blood tests were performed in the morning on the second hospitalization day, from a forearm vein after at least 10 hours of fasting.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003ePredictors Selection\u003c/h2\u003e \u003cp\u003ePredictor selection was primarily performed using traditional statistical method and machine learning approaches. Firstly, an initial variable screening was performed using univariable analysis [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], and only covariates with a \u003cem\u003ep\u003c/em\u003e-value of less than 0.01 were chosen for subsequent analysis. Secondly, predictors were further selected from the above variables using the best subset selection method, via the leaps package (Version 3.1) with complete cases [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. During the process, we set the maximum size of the subset to eight, which was also the default number in the function. During repeated iterations, information criterions including Mallows\u0026rsquo; Cp (CP), and Bayes Information Criteria (BIC) of different subset sizes were demonstrated in plots, all of which help determining the best subset [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Thirdly, decision curve analysis (DCA) was used to choose the final model when different information criterion directed to different best subsets [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eDevelopment of Prediction Model\u003c/h2\u003e \u003cp\u003eThe prediction model was developed using logistic regression, and variables were excluded if their coefficients became insignificant after adjusting for the psychotropic medication use. Since variables were selected without considering observations with missing data, we used multiple imputation by chained equations (MICE) to avoid bias or inefficient estimates of parameters [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. All results of the blood tests, in addition to age and gender, as well as the dichotomous outcome variable were included in the imputation. With the assumption that data were missing at random (MAR), the predictive mean matching (PMM) method was used to impute the missing variables using the mice package (version 3.15.0) in R.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003ePrediction Model Presentation and Examination\u003c/h2\u003e \u003cp\u003eThe prediction model was presented in the form of a nomogram and its accuracy was tested using the receiver operating characteristics (ROC) analysis [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e], after deleting cases with missing data of the selected variables in the original study population. The area under the curve (AUC) with a higher value indicated higher accuracy. The model would be considered superior to a random ordering model if AUC\u0026thinsp;\u0026gt;\u0026thinsp;0.5, while AUC value ranging from 0.5\u0026ndash;0.7 indicate mild performance, 0.7\u0026ndash;0.9 indicate moderate performance. In addition, sensitivity, specificity, and the likelihood ratios [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e], including the positive likelihood ratio (LR [+]) and the negative likelihood ratio (LR [-]) were calculated to further test the accuracy of the model. After verifying the multivariable logistic model, we built a corresponding nomogram to provide a quantitative tool to help distinguish patients with MDD or BD.\u003c/p\u003e \u003cp\u003eCalibration curves were plotted to assess the calibration of the nomogram [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], with the final regression model subjected to bootstrapping validation (1,000 bootstrap resamples), via the rms package (Version 6.3-0).\u003c/p\u003e \u003cp\u003eMoreover, internal validation was performed using 10-fold cross-validation repeating 10 times [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], via caret package (Version 6.0\u0026ndash;93). Moreover, different subsets were used to further validate the model, including the drug na\u0026iuml;ve subset, and the different age subgroups including 14\u0026ndash;29 years, 30\u0026ndash;44 years, and 45\u0026thinsp;+\u0026thinsp;years.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analysis\u003c/h2\u003e \u003cp\u003eData for continuous variables are presented as mean and standard deviation (SD), skewed data as median (25th and 75th percentiles), and categorical variables as absolute numbers and percentages. Clinical characteristics were compared using the Wilcoxon rank sum test for continuous variables and Pearson\u0026rsquo;s Chi-squared test, Fisher\u0026rsquo;s exact test when required, for categorical variables. All statistical analyses were conducted using the freely available statistical software R (version 4.2.0). The reported statistical significance levels were all two-sided, with an alpha value set at 0.05.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\n \u003ch2\u003eEpidemiological Characteristics\u003c/h2\u003e\n \u003cp\u003eIn total, 721 patients were included in the current study, 234 in the MDD group and 487 in the BD group. Characteristics of the study population are given in Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e. There were no statistically significant differences between patients with MDD and BD in gender, duration of illness, and family history of mental disorders, while patients of the two groups had different features in age, marital status, and use rate of antipsychotics, antidepressants, and mood stabilizers (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01).\u003c/p\u003e\u0026nbsp;\u003ctable id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eEpidemiological characteristics of patients with MDD or BP\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eVariables\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eOverall,\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;721\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eMDD,\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;234\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eBP,\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;487\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAge, years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e21 (16, 37)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e47 (27, 57)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e17 (15, 24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGender, No. (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.061\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eFemale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e537 (74%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e164 (70%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e373 (77%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e184 (26%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e70 (30%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e114 (23%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDuration, years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (1, 4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (0.5, 5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (1, 4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eFamily history, No. (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e120 (17%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e38 (16%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e82 (17%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMarital status, No. (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDivorced\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e20 (2.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e8 (3.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e12 (2.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMarried\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e223 (31%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e150 (64%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e73 (15%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSingle\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e474 (66%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e72 (31%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e402 (83%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eWidowed\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4 (0.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4 (1.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0 (0.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAntipsychotics use, No. (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e332 (46%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e68 (29%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e264 (54%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAntidepressants use, No. (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e278 (39%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e124 (53%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e154 (32%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMood stabilizers use, No. (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e277 (38%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e19 (8.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e258 (53%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBenzodiazepines use, No. (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e284 (39%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e101 (43%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e183 (38%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"5\"\u003e\n \u003cp\u003eData are presented as median (IQR) or a number (percentage). \u003cem\u003eP\u003c/em\u003e value is derived from univariable analyses using Wilcoxon rank sum test or Pearson\u0026apos;s Chi-squared test (Fisher\u0026apos;s exact test when needed). The psychotropic medication uses were limited to a month preceding study entry.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003cp\u003e\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\n \u003ch2\u003eVariable Selection\u003c/h2\u003e\n \u003cp\u003eConcerning results of blood tests, 226 (31.34%) participants had different degrees of data missing, most of which were concentrated on the examinations of the inflammatory and immune response (Supplement Figure). We did not delete incomplete observations directly but used complete cases for primary variable selection and then imputed incomplete cases for further validation.\u003c/p\u003e\n \u003cp\u003eWith preliminary data screening using univariable analysis, 22 potential biomarkers including age with a \u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;\u0026lt;\u0026thinsp;0.01 were selected for best subset selection (Supplement Table). As demonstrated (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e), the subset with eight variables showed the smallest CP (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eA). However, the regression coefficient for platelet-to-lymphocyte ratio (P.L) was not significant (p\u0026thinsp;\u0026cong;\u0026thinsp;0.09), so we decided to delete P.L from the model after verifying that its exclusion did not make a significant statistic difference using the ANOVA function in R. The subset with five variables showed the smallest BIC (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eB). According to different information criterions, CP or BIC, a model with 7 or 5 specific variables could arguably be the best subset. Then how could we decide which model to choose? Here, we used DCA to compare the net benefit from the two diagnosis models, and it was shown that the latter one had moderately greater clinical benefits in general (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eC), which consisted of age (unit: years), eosinophil count (Eos, unit: 10\u003csup\u003e9\u003c/sup\u003e/L), plasma concentrations of thyroid-stimulating hormone (TSH, unit: uIU/mL), follicle-stimulating hormone (FSH, unit: mIU/mL), prolactin (PRL, unit: ng/mL), total cholesterol (TC, unit: mmol/L), and low-density lipoprotein cholesterol (LDL, unit: mmol/L).\u003c/p\u003e\n \u003cp\u003eSince complete case analysis may introduce bias, we used imputed datasets for consistency checks. If the conclusions drawn from the observatory cases or imputed cases were consistent, we could be confident that the conclusions were reliable. In addition, we briefly examined whether the present psychotropic medication use, including antipsychotics, antidepressants, mood stabilizers, and benzodiazepines, would influence the selected variables. It was shown that complete cases and imputed datasets provided similar results from multivariable logistic regression analysis (Table \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e). After adjusting for present use of some psychotropic medications, the coefficients (\u0026beta;) and odd ratios (or exp(\u0026beta;)) of TSH and FSH became insignificant, which indicated these two indices were notably affected by the medications. Thus, we deleted TSH and FSH from the final model.\u003c/p\u003e\u0026nbsp;\u003ctable id=\"Tab2\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eMultivariable regression for diagnosis between MDD and BP in patients\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" rowspan=\"2\" style=\"width: 15.2621%;\"\u003e\n \u003cp\u003eIntercept and variables\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"3\" style=\"width: 21.9534%;\"\u003e\n \u003cp\u003eModel of complete cases\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"3\" style=\"width: 21.9534%;\"\u003e\n \u003cp\u003eModel of multiple imputations\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e\u0026beta;\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" style=\"width: 12.1044%;\"\u003e\n \u003cp\u003eOdds Ratio\u003c/p\u003e\n \u003cp\u003e(95% CI)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e\u003cem\u003eP\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e\u0026beta;\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" style=\"width: 12.1044%;\"\u003e\n \u003cp\u003eOdds Ratio\u003c/p\u003e\n \u003cp\u003e(95% CI)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e\u003cem\u003eP\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 15.2621%;\"\u003e\n \u003cp\u003eIntercept\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e2.539\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 12.1044%;\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e2.587\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 12.1044%;\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 15.2621%;\"\u003e\n \u003cp\u003eEOS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e2.358\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 12.1044%;\"\u003e\n \u003cp\u003e10.6 (1.69 to 75.17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e0.015\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e2.392\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 12.1044%;\"\u003e\n \u003cp\u003e10.9 (1.66 to 72.23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e0.013\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 15.2621%;\"\u003e\n \u003cp\u003eTSH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e0.102\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 12.1044%;\"\u003e\n \u003cp\u003e1.10 (0.99 to 1.26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e0.104\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e0.104\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 12.1044%;\"\u003e\n \u003cp\u003e1.11 (0.98 to 1.25)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e0.098\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 15.2621%;\"\u003e\n \u003cp\u003eFSH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e-0.014\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 12.1044%;\"\u003e\n \u003cp\u003e0.99 (0.97 to 1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e0.068\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e-0.013\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 12.1044%;\"\u003e\n \u003cp\u003e0.99 (0.97 to 1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e0.091\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 15.2621%;\"\u003e\n \u003cp\u003ePRL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e0.012\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 12.1044%;\"\u003e\n \u003cp\u003e1.01 (1.00 to 1.02)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e0.006\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e0.012\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 12.1044%;\"\u003e\n \u003cp\u003e1.01 (1.00 to 1.02)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e0.004\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 15.2621%;\"\u003e\n \u003cp\u003eLDL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e0.873\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 12.1044%;\"\u003e\n \u003cp\u003e2.39 (1.29 to 4.54)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e0.006\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e0.882\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 12.1044%;\"\u003e\n \u003cp\u003e2.42 (1.30 to 4.49)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e0.005\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 15.2621%;\"\u003e\n \u003cp\u003eTC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e-0.698\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 12.1044%;\"\u003e\n \u003cp\u003e0.50 (0.30 to 0.82)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e0.007\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e-0.711\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 12.1044%;\"\u003e\n \u003cp\u003e0.49 (0.30 to 0.81)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e0.005\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 15.2621%;\"\u003e\n \u003cp\u003eage\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e-0.067\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 12.1044%;\"\u003e\n \u003cp\u003e0.93 (0.92 to 0.95)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.6613%;\"\u003e\n \u003cp\u003e-0.072\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 12.1044%;\"\u003e\n \u003cp\u003e0.93 (0.91 to 0.95)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 5.1876%;\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"7\" style=\"width: 60.0711%;\"\u003e\n \u003cp\u003e\u0026beta; is the regression coefficient. Models are presented after adjustment of use of antipsychotics, antidepressants, mood stabilizers, and benzodiazepines.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003cp\u003e\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\n \u003ch2\u003ePresentation of Prediction Model\u003c/h2\u003e\n \u003cp\u003eMultivariable logistic regression analysis identified that age, LDL, TC, Eos, and PRL are independent predictors. The model incorporating these independent predictors was developed and presented as a nomogram (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\n \u003ch2\u003eEvaluation of Model Performance\u003c/h2\u003e\n \u003cp\u003eModel performance includes two characteristics, discrimination and calibration. Discrimination refers to the ability to distinguish between the two outcomes and can be assessed by concordance statistic (c-statistic). In logistic regression analysis, the c-statistic is equal to the area under the ROC curve (AUC). For the above nomogram, the c-statistic was 0.858, indicating good discrimination (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eA). Moreover, a cutoff value of 0.66 showed a sensitivity of 0.716 and a specificity of 0.890. These values were calculated using observatory data containing 700 cases without missing values of selected variables. Moreover, LR [+] and LR [-] were 6.51 and 0.32, suggesting moderate shifts in probability of a correct diagnosis using the model.\u003c/p\u003e\n \u003cp\u003eCalibration is used to evaluate the goodness of fit of the prediction model. The calibration plot here indicated that predicted probabilities approximately matched actual probabilities (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eB). For this model, the Hosmer-Lemeshow test \u003cem\u003ep\u003c/em\u003e-value was 0.705, indicating no evidence of poor model fit, in other words, the model seemed to fit well.\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eValidation of Prediction\u003c/strong\u003e \u003cstrong\u003eModel\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eTen-fold cross validation was used for internal validation. The process of cross was repeated ten times (100 samples in total), and the average c-statistic, or AUC, across the samples was 0.853 (range from 0.850 to 0.856) (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eA). This was close to but slightly lower than the overall model c-statistic of 0.858. The values showed the stability and reliability of predictions within the study population, and indicated good performance when the model is applied to new patients. Moreover, subset validation with ROC curve was used to further evaluate the effectiveness of the model. In the drug na\u0026iuml;ve subset, the AUC was 0.826, indicating good discrimination (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eB). In different age subsets, the AUC ranged from 0.671 to 0.739, indicating mild to moderate discrimination (Fig. 4C1-C3).\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eAfter many years of effort, researchers have not yet constructed a prediction model for discriminating between BD and MDD with clinical utility. In the present study, we preliminarily developed and validated a diagnostic nomogram, with a composite of biomarkers from routinely tested blood results, to distinguish MDD and BD. The model was constructed using the best subset selection method and then verified and adjusted using multiple imputations and the inclusion of psychotropic medication use. The final model consisted of five variables: age, LDL, TC, Eos, and PRL. The model could discriminate between MDD and BD with an AUC of 0.858, with a sensitivity of 0.716 and a specificity of 0.890.\u003c/p\u003e \u003cp\u003eDuring the construction of the model, 47 features were reduced to 22 potential predictors at the first step by univariant analysis, then the best subset selection method was managed to select seven prominent markers. Of the 721 patients in the study, only 495 without missing data were used for the primary multivariable selections. Then 700 were used for adjustment and evaluation of the model after deleting cases with incomplete values of the prominent variables, which made the findings relatively more robust than constructing and validating the model using the same population. Moreover, repeated cross validations were subsequently used to verify the model when the training dataset and test dataset did not overlap, and subset validations were used to test the effectiveness of the model in drug na\u0026iuml;ve patients and patients of different age groups.\u003c/p\u003e \u003cp\u003eThe findings of the present study were somewhat consistent with previous studies. For example, age is one of the most profound distinguishing factors between MDD and BD, as it had been broadly accepted that the onset age of MDD is generally later than that of BD [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. However, we wanted to see how the performance of the composite biomarkers would change if the effect of age was minimized. The study divided patients into three groups, 14\u0026ndash;29 years, 30\u0026ndash;44 years group, and 45\u0026thinsp;+\u0026thinsp;years respectively. Within each group, age became insignificant different between MDD and BP patients (data not shown). Unsurprisingly, the model discrimination had varying degrees of deteriorations, and the AUC were 0.688, 0.671, and 0.739 respectively, indicating that the model still had mild to moderate diagnostic efficiency in patients of same age group.\u003c/p\u003e \u003cp\u003eMoreover, eosinophil could also help discriminate the two disorders, which was consistent with previous studies. For example, it has been demonstrated that eosinophil counts were reduced in MDD patients [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], while the increased eosinophil function could be found in the late-stage of BD [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn addition, the inclusion of PRL in the model, one of the hormones secreted by pituitary gland, suggested that pituitary function might play a role in differentiating MDD and BD. However, previous studies on the pituitary gland mainly focused on the gland volume changes in mental disorders and their association with hyperactivity in the Hypothalamic-Pituitary-Adrenal axis [\u003cspan additionalcitationids=\"CR20\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Other hormones provided by pituitary gland besides adrenocorticotropic hormone could also have potential effects on mental disorders. In this study, TSH and FSH were tested statistically significant but excluded after preliminary adjustment for psychotropic medication use, which was in accordance with clinical consensus that endocrine is greatly influenced during the drug treatment for affective disorders [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Interestingly, PRL remained in the model after medication adjustment. However, these findings require further confirmation in drug-free patients.\u003c/p\u003e \u003cp\u003eIn addition, LDL and TC were also included in the final model. These findings did not contradict previous findings that abnormal lipid metabolism was more prevalent in MDD and BD patients compared to health controls [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. However, few studies have compared the differences in lipid profile distribution between MDD and BD. Our study showed that BD patients had relatively higher LDL levels, while MDD patients had higher TC levels. Although these findings indicated different lipid profiles in MDD and BP patients, but both were consistent with the findings that patients with severe mental illnesses had increased risks for cardiovascular diseases [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eMoreover, like endocrine functions, lipid metabolism is also seriously affected by some kinds of psychotropic drugs, especially antipsychotics and mood stabilizers, such as clozapine, olanzapine, and valproate [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e], which can ultimately result in hyperlipidemia or even obesity. As it was demonstrated in Table \u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, the proportion of BD patients using antipsychotics and mood stabilizers was significantly higher than that of MDD patients, however, the coefficients of TC and LDL in the regression model remained significant after the adjustment of medication use in this study, indicating that the pharmacological effect was not the only reason for the differences in the lipid levels between the two groups. In other words, abnormal lipid metabolism may underlie the mental disorders. However, since the cholesterol level can be greatly influenced by living habits, such as diet and physical activities [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e], the significance could not be applied to a population with different lifestyles.\u003c/p\u003e \u003cp\u003eEmerging studies have confirmed the potential roles of inflammation or immune-based biomarkers as predictive biomarker panels to differentiate MDD and BD, usually including C-reactive protein (CRP), interleukins, and complement components [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan additionalcitationids=\"CR30\" citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. However, the above related potential biomarkers were surprisingly excluded during model development, which was inconsistent with previous findings. For example, Chang et.al demonstrated that baseline C-reactive protein could serve as a discrimination biomarker for MDD and bipolar II disorder in drug na\u0026iuml;ve patients (cutoff value: 621.6 ng/mL; AUC value: 0.816), and patients with baseline C-reactive protein greater than 621.6 ng/mL had 28.2 higher odds of bipolar II disorder [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. However, in our study, C-reactive protein level showed no statistical difference between MDD and BD and was excluded at the first step. The possible reason might be treatment effects as indicated by Chang\u0026rsquo;s work itself: the difference of C-reactive protein level would become narrower between MDD and bipolar II disorder after treatment. Another possible reason may be bias from concentrated missing values on inflammation and immune factors; although the multiple imputations had indicated that the missingness of the selected variables in the model was at random, it may not represent the same missing pattern of the potential predictors in question [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThere were several limitations to this study. Firstly, behavior characteristics and psychological assessments failed to be included in the analysis process. Secondly, BD patients were not specifically classified into different clinical phases including (hypo)manic or depressive phase, or rapid cycling BD. Moreover, although the diagnostic biomarker panel constructed by this study was preliminarily unlikely to be affected by medications as cofounding factors but should be tested further in drug na\u0026iuml;ve patients. At last, the data were collected from one hospital, the generalizability of the prediction panel needs further testing.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eOur study presents a nomogram that incorporates factors from commonly tested blood indicators that can conveniently help distinguishing MDD and BD, and thus reduce misdiagnosis.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eJieping Huang designed the study, analyzed the data, and wrote the manuscript. Xuejiao Hou, Moyan Li, Yingshuang Xue collected the data. Jiangfei An, Shenglin Wen and Zi Wang reviewed the writing of the manuscript. Minfeng Cheng and Jihui Yue supervised the study. All authors contributed to and have approved the submitted manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of Data and Materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data of this study cannot be made publicly available for confidentiality reasons. Data are however available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003eAcknowledgement\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003eCompeting interests\u003c/p\u003e\n\u003cp\u003eThe authors declare no financial or other conflict of interest.\u003c/p\u003e\n\u003cp\u003eConsent for publication\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003eEthical approval and consent to participate\u003c/p\u003e\n\u003cp\u003eThe research was conducted according to the International Conference on Harmonization \u0026ndash; Good Clinical Practice Guidelines and the Declaration of Helsinki. The study procedures were reviewed and approved by the Medical Ethic Committee Review Board of the Fifth Affiliated Hospital of Sun Yat-sen University (Approval number: [2022] K44-1). The need for informed consent was waived by the Medical Ethic Committee Review Board of the Fifth Affiliated Hospital of Sun Yat-sen University. This was because the research used human materials or data with identifiable personal information, but the subject could no longer be found and the research project did not involve personal privacy or commercial interests, according to the Article 39 of the \u0026lsquo;Ethical Review Measures for Biomedical Research Involving Humans\u0026rsquo; issued by the National Health and Family Planning Commission of the People\u0026rsquo;s Republic of China.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eN. H. Kalin, \u0026quot;New Insights Into Major Depression and the Treatment of Bipolar Depression,\u0026quot; (in eng), \u003cem\u003eAm J Psychiatry, \u003c/em\u003evol. 178, no. 12, pp. 1071-1074, Dec 2021, doi: 10.1176/appi.ajp.2021.21101042.\u003c/li\u003e\n\u003cli\u003eI. Grande, M. Berk, B. Birmaher, and E. Vieta, \u0026quot;Bipolar disorder,\u0026quot; (in eng), \u003cem\u003eLancet, \u003c/em\u003evol. 387, no. 10027, pp. 1561-1572, Apr 09 2016, doi: 10.1016/S0140-6736(15)00241-X.\u003c/li\u003e\n\u003cli\u003eE. A. Stahl\u003cem\u003e et al.\u003c/em\u003e, \u0026quot;Genome-wide association study identifies 30 loci associated with bipolar disorder,\u0026quot; (in eng), \u003cem\u003eNat Genet, \u003c/em\u003evol. 51, no. 5, pp. 793-803, May 2019, doi: 10.1038/s41588-019-0397-8.\u003c/li\u003e\n\u003cli\u003eL. Fusar-Poli\u003cem\u003e et al.\u003c/em\u003e, \u0026quot;Neutrophil-to-Lymphocyte, Platelet-to-Lymphocyte and Monocyte-to-Lymphocyte Ratio in Bipolar Disorder,\u0026quot; (in eng), \u003cem\u003eBrain Sci, \u003c/em\u003evol. 11, no. 1, Jan 06 2021, doi: 10.3390/brainsci11010058.\u003c/li\u003e\n\u003cli\u003eB. I. Perry, R. Upthegrove, N. Kappelmann, P. B. Jones, S. Burgess, and G. M. Khandaker, \u0026quot;Associations of immunological proteins/traits with schizophrenia, major depression and bipolar disorder: A bi-directional two-sample mendelian randomization study,\u0026quot; (in eng), \u003cem\u003eBrain Behav Immun, \u003c/em\u003evol. 97, pp. 176-185, Oct 2021, doi: 10.1016/j.bbi.2021.07.009.\u003c/li\u003e\n\u003cli\u003eK. Munkholm, M. Vinberg, B. K. Pedersen, H. E. Poulsen, C. T. Ekstr\u0026oslash;m, and L. V. Kessing, \u0026quot;A multisystem composite biomarker as a preliminary diagnostic test in bipolar disorder,\u0026quot; (in eng), \u003cem\u003eActa Psychiatr Scand, \u003c/em\u003evol. 139, no. 3, pp. 227-236, Mar 2019, doi: 10.1111/acps.12983.\u003c/li\u003e\n\u003cli\u003eS. J. Rhee\u003cem\u003e et al.\u003c/em\u003e, \u0026quot;Comparison of serum protein profiles between major depressive disorder and bipolar disorder,\u0026quot; (in eng), \u003cem\u003eBMC Psychiatry, \u003c/em\u003evol. 20, no. 1, p. 145, Apr 03 2020, doi: 10.1186/s12888-020-02540-0.\u003c/li\u003e\n\u003cli\u003eE. H. Au, A. Francis, A. Bernier-Jean, and A. Teixeira-Pinto, \u0026quot;Prediction modeling-part 1: regression modeling,\u0026quot; (in eng), \u003cem\u003eKidney Int, \u003c/em\u003evol. 97, no. 5, pp. 877-884, May 2020, doi: 10.1016/j.kint.2020.02.007.\u003c/li\u003e\n\u003cli\u003eG. Heinze, C. Wallisch, and D. Dunkler, \u0026quot;Variable selection - A review and recommendations for the practicing statistician,\u0026quot; (in eng), \u003cem\u003eBiom J, \u003c/em\u003evol. 60, no. 3, pp. 431-449, May 2018, doi: 10.1002/bimj.201700067.\u003c/li\u003e\n\u003cli\u003eR. I. Kabacoff, \u003cem\u003eR in Action\u003c/em\u003e, 2 ed. New York, NY: Manning Publications, 2015.\u003c/li\u003e\n\u003cli\u003eM. Fitzgerald, B. R. Saville, and R. J. Lewis, \u0026quot;Decision curve analysis,\u0026quot; (in eng), \u003cem\u003eJAMA, \u003c/em\u003evol. 313, no. 4, pp. 409-10, Jan 27 2015, doi: 10.1001/jama.2015.37.\u003c/li\u003e\n\u003cli\u003eK. Blazek, A. van Zwieten, V. Saglimbene, and A. Teixeira-Pinto, \u0026quot;A practical guide to multiple imputation of missing data in nephrology,\u0026quot; (in eng), \u003cem\u003eKidney Int, \u003c/em\u003evol. 99, no. 1, pp. 68-74, Jan 2021, doi: 10.1016/j.kint.2020.07.035.\u003c/li\u003e\n\u003cli\u003eM. Weatherall, \u0026quot;Information provided by diagnostic and screening tests: improving probabilities,\u0026quot; (in eng), \u003cem\u003ePostgrad Med J, \u003c/em\u003evol. 94, no. 1110, pp. 230-235, Apr 2018, doi: 10.1136/postgradmedj-2017-135273.\u003c/li\u003e\n\u003cli\u003eY. Q. Huang\u003cem\u003e et al.\u003c/em\u003e, \u0026quot;Development and Validation of a Radiomics Nomogram for Preoperative Prediction of Lymph Node Metastasis in Colorectal Cancer,\u0026quot; (in eng), \u003cem\u003eJ Clin Oncol, \u003c/em\u003evol. 34, no. 18, pp. 2157-64, Jun 20 2016, doi: 10.1200/JCO.2015.65.9128.\u003c/li\u003e\n\u003cli\u003eE. Vieta\u003cem\u003e et al.\u003c/em\u003e, \u0026quot;Early Intervention in Bipolar Disorder,\u0026quot; (in eng), \u003cem\u003eAm J Psychiatry, \u003c/em\u003evol. 175, no. 5, pp. 411-426, May 01 2018, doi: 10.1176/appi.ajp.2017.17090972.\u003c/li\u003e\n\u003cli\u003eL. T. Park and C. A. Zarate, \u0026quot;Depression in the Primary Care Setting,\u0026quot; (in eng), \u003cem\u003eN Engl J Med, \u003c/em\u003evol. 380, no. 6, pp. 559-568, Feb 07 2019, doi: 10.1056/NEJMcp1712493.\u003c/li\u003e\n\u003cli\u003eD. Singh\u003cem\u003e et al.\u003c/em\u003e, \u0026quot;Changes in leukocytes and CRP in different stages of major depression,\u0026quot; (in eng), \u003cem\u003eJ Neuroinflammation, \u003c/em\u003evol. 19, no. 1, p. 74, Apr 04 2022, doi: 10.1186/s12974-022-02429-7.\u003c/li\u003e\n\u003cli\u003eB. Panizzutti\u003cem\u003e et al.\u003c/em\u003e, \u0026quot;Increased serum levels of eotaxin/CCL11 in late-stage patients with bipolar disorder: An accelerated aging biomarker?,\u0026quot; (in eng), \u003cem\u003eJ Affect Disord, \u003c/em\u003evol. 182, pp. 64-9, Aug 15 2015, doi: 10.1016/j.jad.2014.12.010.\u003c/li\u003e\n\u003cli\u003eG. Delvecchio\u003cem\u003e et al.\u003c/em\u003e, \u0026quot;Pituitary gland shrinkage in bipolar disorder: The role of gender,\u0026quot; (in eng), \u003cem\u003eCompr Psychiatry, \u003c/em\u003evol. 82, pp. 95-99, Apr 2018, doi: 10.1016/j.comppsych.2018.01.014.\u003c/li\u003e\n\u003cli\u003eG. Delvecchio, A. C. Altamura, J. C. Soares, and P. Brambilla, \u0026quot;Pituitary gland in Bipolar Disorder and Major Depression: Evidence from structural MRI studies: Special Section on \u0026quot;Translational and Neuroscience Studies in Affective Disorders\u0026quot;. Section Editor, Maria Nobile MD, PhD. This Section of JAD focuses on the relevance of translational and neuroscience studies in providing a better understanding of the neural basis of affective disorders. The main aim is to briefly summarise relevant research findings in clinical neuroscience with particular regards to specific innovative topics in mood and anxiety disorders,\u0026quot; (in eng), \u003cem\u003eJ Affect Disord, \u003c/em\u003evol. 218, pp. 446-450, Aug 15 2017, doi: 10.1016/j.jad.2017.03.066.\u003c/li\u003e\n\u003cli\u003eJ. Keller\u003cem\u003e et al.\u003c/em\u003e, \u0026quot;HPA axis in major depression: cortisol, clinical symptomatology and genetic variation predict cognition,\u0026quot; (in eng), \u003cem\u003eMol Psychiatry, \u003c/em\u003evol. 22, no. 4, pp. 527-536, Apr 2017, doi: 10.1038/mp.2016.120.\u003c/li\u003e\n\u003cli\u003eJ. F. Hayes, L. Marston, K. Walters, J. R. Geddes, M. King, and D. P. Osborn, \u0026quot;Adverse Renal, Endocrine, Hepatic, and Metabolic Events during Maintenance Mood Stabilizer Treatment for Bipolar Disorder: A Population-Based Cohort Study,\u0026quot; (in eng), \u003cem\u003ePLoS Med, \u003c/em\u003evol. 13, no. 8, p. e1002058, Aug 2016, doi: 10.1371/journal.pmed.1002058.\u003c/li\u003e\n\u003cli\u003eY. G. Wei\u003cem\u003e et al.\u003c/em\u003e, \u0026quot;Cholesterol and triglyceride levels in first-episode patients with major depressive disorder: A meta-analysis of case-control studies,\u0026quot; \u003cem\u003eJ Affect Disord, \u003c/em\u003evol. 266, pp. 465-472, Apr 1 2020, doi: 10.1016/j.jad.2020.01.114.\u003c/li\u003e\n\u003cli\u003eN. Dalkner\u003cem\u003e et al.\u003c/em\u003e, \u0026quot;Metabolic Syndrome Impairs Executive Function in Bipolar Disorder,\u0026quot; (in eng), \u003cem\u003eFront Neurosci, \u003c/em\u003evol. 15, p. 717824, 2021, doi: 10.3389/fnins.2021.717824.\u003c/li\u003e\n\u003cli\u003eB. I. Goldstein\u003cem\u003e et al.\u003c/em\u003e, \u0026quot;Major Depressive Disorder and Bipolar Disorder Predispose Youth to Accelerated Atherosclerosis and Early Cardiovascular Disease: A Scientific Statement From the American Heart Association,\u0026quot; (in eng), \u003cem\u003eCirculation, \u003c/em\u003evol. 132, no. 10, pp. 965-86, Sep 08 2015, doi: 10.1161/CIR.0000000000000229.\u003c/li\u003e\n\u003cli\u003eR. E. Nielsen, J. Banner, and S. E. Jensen, \u0026quot;Cardiovascular disease in patients with severe mental illness,\u0026quot; (in eng), \u003cem\u003eNat Rev Cardiol, \u003c/em\u003evol. 18, no. 2, pp. 136-145, Feb 2021, doi: 10.1038/s41569-020-00463-7.\u003c/li\u003e\n\u003cli\u003eT. Pillinger\u003cem\u003e et al.\u003c/em\u003e, \u0026quot;Comparative effects of 18 antipsychotics on metabolic function in patients with schizophrenia, predictors of metabolic dysregulation, and association with psychopathology: a systematic review and network meta-analysis,\u0026quot; (in eng), \u003cem\u003eLancet Psychiatry, \u003c/em\u003evol. 7, no. 1, pp. 64-77, Jan 2020, doi: 10.1016/S2215-0366(19)30416-X.\u003c/li\u003e\n\u003cli\u003eU. E. Bauer, P. A. Briss, R. A. Goodman, and B. A. Bowman, \u0026quot;Prevention of chronic disease in the 21st century: elimination of the leading preventable causes of premature death and disability in the USA,\u0026quot; (in eng), \u003cem\u003eLancet, \u003c/em\u003evol. 384, no. 9937, pp. 45-52, Jul 05 2014, doi: 10.1016/S0140-6736(14)60648-6.\u003c/li\u003e\n\u003cli\u003eG. R. Fries, M. J. Zamzow, T. Andrews, O. Pink, G. Scaini, and J. Quevedo, \u0026quot;Accelerated aging in bipolar disorder: A comprehensive review of molecular findings and their clinical implications,\u0026quot; (in eng), \u003cem\u003eNeurosci Biobehav Rev, \u003c/em\u003evol. 112, pp. 107-116, May 2020, doi: 10.1016/j.neubiorev.2020.01.035.\u003c/li\u003e\n\u003cli\u003eF. Haenisch\u003cem\u003e et al.\u003c/em\u003e, \u0026quot;Towards a blood-based diagnostic panel for bipolar disorder,\u0026quot; (in eng), \u003cem\u003eBrain Behav Immun, \u003c/em\u003evol. 52, pp. 49-57, Feb 2016, doi: 10.1016/j.bbi.2015.10.001.\u003c/li\u003e\n\u003cli\u003eH. Yu\u003cem\u003e et al.\u003c/em\u003e, \u0026quot;Association of the plasma complement system with brain volume deficits in bipolar and major depressive disorders,\u0026quot; (in eng), \u003cem\u003ePsychol Med, \u003c/em\u003epp. 1-11, Oct 26 2022, doi: 10.1017/S0033291722003282.\u003c/li\u003e\n\u003cli\u003eH. H. Chang\u003cem\u003e et al.\u003c/em\u003e, \u0026quot;C-reactive protein: A differential biomarker for major depressive disorder and bipolar II disorder,\u0026quot; (in eng), \u003cem\u003eWorld J Biol Psychiatry, \u003c/em\u003evol. 18, no. 1, pp. 63-70, Feb 2017, doi: 10.3109/15622975.2016.1155746.\u003c/li\u003e\n\u003cli\u003eI. R. White, P. Royston, and A. M. Wood, \u0026quot;Multiple imputation using chained equations: Issues and guidance for practice,\u0026quot; (in eng), \u003cem\u003eStat Med, \u003c/em\u003evol. 30, no. 4, pp. 377-99, Feb 20 2011, doi: 10.1002/sim.4067.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-psychiatry","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bpsy","sideBox":"Learn more about [BMC Psychiatry](http://bmcpsychiatry.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bpsy/default.aspx","title":"BMC Psychiatry","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Major depressive disorder, Bipolar disorder, Diagnosis, Nomogram, Validation","lastPublishedDoi":"10.21203/rs.3.rs-3058571/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3058571/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eObjective\u003c/h2\u003e \u003cp\u003eThe purpose of the present study was to construct a diagnostic model to distinguish major depressive disorder (MDD) and bipolar disorder (BD) using potential commonly tested blood biomarkers.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eInformation of 721 inpatients with an ICD-10 diagnosis of major depressive disorder or bipolar disorder were collected from the electronic medical record system. Variables in the nomogram were selected by best subset selection method after a prior univariable screening, and then constructed using logistic regression with inclusion of the psychotropic medication use. The discrimination, calibration and internal validation of the nomogram were evaluated by the receiver operating characteristic curve, the calibration curve, cross validation and subset validation method.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThe nomogram consisted of five variables, including age, eosinophil count, plasma concentrations of prolactin, total cholesterol, and low-density lipoprotein cholesterol. The model could discriminate between MDD and BD with an AUC of 0.858, with a sensitivity of 0.716 and a specificity of 0.890.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eThe comprehensive nomogram constructed by the present study can be convenient to distinguish MDD and BD since the incorporating variables were common indicators in clinical practice. It could help avoid misdiagnoses and improve prognosis of the patients.\u003c/p\u003e","manuscriptTitle":"A composite of blood-based biomarkers to distinguish major depressive disorder and bipolar disorder in adolescents and adults","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2023-06-26 23:17:26","doi":"10.21203/rs.3.rs-3058571/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Major revision","date":"2023-08-21T10:40:27+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2023-08-15T09:04:45+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"548f0951-32b2-443b-bca3-a91de2e14af6","date":"2023-08-14T23:59:11+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"d6260c94-6389-4736-b78b-0e9f82cec3e3","date":"2023-08-13T16:51:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"33e0b464-be24-4810-8478-b8d9039f5ea1","date":"2023-08-11T16:49:50+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2023-07-04T19:52:45+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2023-07-04T19:51:31+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2023-06-22T04:33:22+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2023-06-22T04:31:55+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Psychiatry","date":"2023-06-13T13:47:38+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-psychiatry","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bpsy","sideBox":"Learn more about [BMC Psychiatry](http://bmcpsychiatry.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bpsy/default.aspx","title":"BMC Psychiatry","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"dc748685-6e56-4a13-92bc-54af0768027d","owner":[],"postedDate":"June 26th, 2023","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2023-10-23T15:03:53+00:00","versionOfRecord":{"articleIdentity":"rs-3058571","link":"https://doi.org/10.1186/s12888-023-05204-x","journal":{"identity":"bmc-psychiatry","isVorOnly":false,"title":"BMC Psychiatry"},"publishedOn":"2023-10-16 15:01:18","publishedOnDateReadable":"October 16th, 2023"},"versionCreatedAt":"2023-06-26 23:17:26","video":"","vorDoi":"10.1186/s12888-023-05204-x","vorDoiUrl":"https://doi.org/10.1186/s12888-023-05204-x","workflowStages":[]},"version":"v1","identity":"rs-3058571","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3058571","identity":"rs-3058571","version":["v1"]},"buildId":"i_I_jhV-QviBjPLXJga9V","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}