A mean platelet volume in inflammatory bowel disease: A systematic review and meta-analysis

Abstract

Background: Inflammatory bowel disease (IBD) is a chronic gastrointestinal tract inflammatory state, which is affecting millions of individuals in the world. It can affect alimentary canals such as colon, rectum, ileum and other parts. In IBD, platelet parameters underwent several changes. Therefore, the aim of this review was determining the estimated pooled mean platelet volume and mean difference in inflammatory bowel disease to elucidate its potential diagnostic value.

Methods: Articles were extensively searched in bibliographic databases using Medical Subject Heading and entry phrases or terms. In addition, articles were directly searched in Google Scholar to account for the studies omission in searching bibliographic databases. Observational (cohort, cross-sectional and case-control) studies, published in English language and conducted on IBD were included. For studies meeting the eligibility criteria, the first author's name, publication year, population, study design, study area, sample size, mean platelet volume and standard deviation were extracted and entered in to Microsoft-excel. The analysis was done by Stata version 11. In order to estimate the pooled mean platelet volume and mean difference, random effect model was done. The heterogeneity was quantified using Higgin's I2 statistics. Publication bias was determined using Egger's test statistics and funnel plot. Sub-group analysis based on population carried to reduce heterogeneity.

Results: A total of 17 relevant articles with 2957 participants (1823 IBD cases and 1134 healthy controls) were included to this study. The pooled estimated MPV was 9.29fl; 95% CI: 9.01-9.57 and 9.50fl; 95% CI: 8.81-10.20 in IBD and control groups, respectively. The standardized pooled estimate of mean difference in mean platelet volume was -0.83fl; 95% CI: -1.15, -0.51; I2: 93.1%; P-value < 0.001. In subgroup analysis based on population, the highest estimated mean difference in MPV was observed among patients of CD; -2.30; 95% CI: -3.46, -1.14; I2: 97.8%; P-value < 0.001.

Conclusion: According to the current systematic review and meta-analysis, mean platelet volume was lower in IBD compared to control. The decreased mean platelet volume could be attributed to platelet consumption or sequestration associated with the progression of IBD. As a result, in IBD, mean platelet volume can provide diagnostic and prognostic information.

Fig 1. PRISMA flow chart describing screening protocols of studies for Meta-analysis.

NB; UC: Ulcerative Colitis; CD: Crohn’s Disease; IBD: Inflammatory Bowel Disease; AA: Acute Appendicitis; CFPIAP: Child Food Protein Induced Allergic Proctocolitis.

Fig 2. Funnel plot showing publication bias of included articles.

Dot on the black line represents each individual article. Y-axis shows the standard error of mean difference (SEMD). The x-axis shows estimate mean difference (MD) of the included articles.

Fig 3. Forest plot of pooled estimated MPV in IBD patients using random effect model.

The size of the x-axis shows the estimate pooled MPV of the studies. In the pooled point calculation, the dotted line represents the MPV. The black dot in the middle of the gray box reflects the estimate pooled MPV of each studies point and the line shows the 95% CI of the estimates. The gray boxes represent each study weight that contributes to the estimation of the pooled MPV. I-squared illustrates the heterogeneity between the included studies.

Fig 4. Forest plot of pooled estimated MPV in healthy controls.

The size of the x-axis shows the estimate of pooled MPV of the studies. MPV is seen in the hard line (MPV = 0). In the pooled point calculation, the dotted line represents the MPV. The black dot in the middle of the gray box reflects the estimate pooled MPV of each studies point and the line shows the 95% CI of the estimates. The gray boxes represent each study weight that contributes to the estimation of the pooled MPV. I-squared illustrates the heterogeneity between the included studies.

Fig 5. Forest plot of pooled estimated SMD using random effect model.

The size of the x-axis shows the SMD estimate of the studies. Hard line indicates no difference (SMD = 0). In the pooled point calculation, the dotted line represents the mean difference. The black dot in the middle of the gray box reflects the SMD estimate of each sample’s point and the line shows the 95% CI of the estimates. The gray boxes represent each study weight that contributes to the estimation of the pooled mean difference. I-squared illustrates the heterogeneity between the included studies.

Fig 6. Forest plot of population based sub-group analysis.

The x-axis scale displays the estimation of the SMD in MPV. The hard line shows no difference. The dotted line represents the mean difference in the pooled point estimate of each study. In the center of the gray box, the black dot reflects the SMD estimate of each article’s point estimate and the line shows the 95% CI of the estimates. I-squared indicates the heterogeneity across the included studies, p indicating for statistical significance of heterogeneity.