Combination of 15 lipid metabolites and motilin to diagnose spleen-deficiency FD

Abstract

Background: This study aims to assess clinical characteristics in FD with spleen deficiency syndrome and metabolic perturbations involved in FD progress. We combined metabolic biomarkers and clinical features into a better prediction for FD with Spleen Deficiency syndrome.

Methods: A total of 276 people were recruited, including 215 FD patients and 61 healthy control group (HC). The clinical characteristics and gastric emptying rate were compared between spleen deficiency-FD group and non-spleen deficiency-FD. The serum lipids metabonomics analysis was performed to determine the metabolic differences in spleen deficiency-FD group and HC.

Results: The symptoms of postprandial discomfort in Spleen Deficiency group were more severe (P < 0.05), and delayed gastric emptying was more pronounced (P < 0.05) vs. non-Spleen deficiency. Decreased motilin (OR = 0.990, 95% confidence interval (CI) 0.982-0.997) was independent risk factor related to Spleen Deficiency group. We identified 15 metabolites for spleen deficiency group vs. HC, majority of those biomarkers belonged to the glycerophospholipid metabolic pathway. The combination of 15 metabolics could diagnose spleen deficiency-FD, with the AUC of 0.9943, 95% CI 0.9854-1.0000), and the combination of 15 metabolics and motilin could diagnose spleen deficiency-FD, with the AUC of 0.9615, 95% CI 0.9264-9967).

Conclusions: This study provides supportive evidence that Spleen deficiency syndrome was associated with delayed gastric emptying and the glycerophospholipid metabolic pathway was perturbed in FD patients. The combination of metabolic biomarkers and clinical features provided us with new ideas for multidimensional diagnosis of FD.Trial registration http://www.chictr.org.cn, no: ChiCTR-TRC-13003200. clinicaltrials.gov, no: NCT02762136.

Keywords: Biomarker; Functional dyspepsia; Metabolomics; Spleen deficiency syndrome.

Conflict of interest statement

The authors declare that they have no competing interests.

Figures

Fig. 1

The flow chart of the selection process

Fig. 2

Postprandial discomfort severity scale (*p 

Fig. 3

Gastric emptying rates of proximal and distal stomach (*p 

Fig. 4

Risk factors for the FD patients with spleen deficiency syndrome. The results of multivariate logistic regression shown that decreased motilin was an independent risk factor related to spleen deficiency in patients with FD

Fig. 5

The PCA scores of spleen deficiency-FD group and healthy control group. (a All spleen deficiency-FD patients and healthy control group; b spleen deficiency-FD patients of XiYuan Hospital and healthy control group; c spleen deficiency-FD patients of GuangDong Province Traditional Chinese Medical Hospital and healthy control group; d spleen deficiency-FD patients of WuHan Integrated TCM &Western Medicine Hospital and healthy control group)

Fig. 6

The OPLS-DA scores of spleen deficiency-FD group and healthy control group. (a spleen deficiency-FD patients of XiYuan Hospital and healthy control group; b spleen deficiency-FD patients of GuangDong Province Traditional Chinese Medical Hospital and healthy control group; c spleen deficiency-FD patients of WuHan Integrated TCM &Western Medicine Hospital and healthy control group)

Fig. 7

The S-Plot of spleen deficiency-FD group and healthy control group. (a spleen deficiency-FD patients of XiYuan Hospital and healthy control group; b spleen deficiency-FD patients of GuangDong Province Traditional Chinese Medical Hospital and healthy control group; c spleen deficiency-FD patients of WuHan Integrated TCM &Western Medicine Hospital and healthy control group)

Fig. 8

Venn plot for all metabonomics biomarkers among three groups. There are 15 overlapping metabolites: LysoPC (18:0), PC (22:6/0:0), LysoPC (17:0), LysoPC (18:1), LysoPC (20:4), PC (17:2/0:0), PC (17:1/0:0), LysoPC (16:1), LysoPE (18:2/0:0), PC (15:0/0:0), PC (16:0/0-1:0), LysoPE (0:0/16:0), LysoPE (20:4/0:0), LysoPE (0:0/22:6), LysoPC (20:3)

Fig. 9

Metabolism map of FD with spleen deficiency syndrome. (1) glycerol phospholipid metabolism; (2) linoleic acid metabolism; (3) sphingolipid metabolism; (4) alpha-linolenic acid metabolism; (5) glyceride metabolism; (6) arachidonic acid metabolism

Fig. 10

ROC curves of biomarkers for Spleen-deficiency in prediction. The prediction of spleen deficiency-FD vs. HC was robust, and the area under the curve (AUC) of 15 biomarkers for spleen deficiency-FD vs. HC ranged from 0.6790 to 0.9926. The combination of 15 biomarkers could offer a highly accurate diagnosis of spleen deficiency in FD with an AUC of 0.9943