Mean Platelet Volume, Platelet Distribution Width, and Platelet Count, in Connection With Immune Thrombocytopenic Purpura and Essential Thrombocytopenia

Eunyup Lee, MD; Miyoung Kim, MD; Kibum Jeon, MD; Jiwon Lee, MD; Jee-Soo Lee, MD; Han-Sung Kim, MD; Hee Jung Kang, MD; Young Kyung Lee, MD


Lab Med. 2019;50(3):279-285. 

In This Article


Previous studies of MPV or PDW focused on the discriminatory power of the 2 indices, mostly in patients with thrombocytopenia and sometimes in those with thrombocytosis. Most of these studies did not compare the values from patients with these disorders with the corresponding values in healthy individuals. In our study, we hypothesized that the MPV and PDW values of circulating PLTs reflect the pathophysiology of these 2 diseases, which have different mechanisms of altering PLT production.

The finding that we believe is most notable is that thrombopoiesis and PLT sequestration in ITP and ET involve complex mechanisms. Patients with ITP showed higher MPV and PDW values than healthy individuals, irrespective of the instrument used. Previous studies have shown that the PDW and/or MPV were higher in patients with ITP than in those with thrombocytopenia of different pathologies, such as hypoproductive thrombocytopenia. Keito et al[14] reported that MPV and PDW were significantly higher in ITP than in aplastic anemia. Ntaios et al[9] reported increased MPV and PDW in patients with ITP, compared with patients with hypoproductive thrombocytopenia. However, the aforementioned studies did not include comparisons to values from healthy individuals. A possible explanation for the higher MPV values in patients with ITP than in healthy individuals is that the proportion of young, large PLTs is higher in ITP than in normal conditions because of compensatory thrombopoiesis in the bone marrow in response to antibody-mediated peripheral destruction.

We expected that PDW values in ITP would be low because we theorized that the continuous destruction of peripheral PLTs would eliminate already existing, relatively old PLTs of smaller sizes; however, our data showed the opposite result. This finding indicates that although the proportion of young, large PLTs is increased in ITP, the high PDW shows that the overall composition of the circulating PLTs varies, with increased proportions of PLTs at both ends of their size distribution. This, in turn, may be occurring because PLTs are not necessarily targeted for destruction because of their older age; rather, the process is nondiscriminatory. Hence, the distribution of the surviving PLTs in patients with ITP is not much different from that in healthy individuals; however, the PDW is higher because of excess younger PLTs. Future investigations of individual PLT sizes and their life spans in patients with ITP ought to yield clarity regarding these notions.

Further, we believed it was notable that the ranges of the distributions of MPV and PDW values were wider in patients with ITP than in those with ET, demonstrating a larger coefficient of variation in the former. Ours is the only study to compare MPV and PDW in ITP and ET; therefore, this finding is novel and could be attributed to the difference in the rates of PLT production in the 2 diseases that have distinct mechanisms for thrombopoiesis (compensatory vs clonal). Compensatory thrombopoiesis in ITP is reactive and therefore, it can be highly sporadic, whereas clonal thrombopoiesis in ET is a relatively stable process.

Another possible explanation for this phenomenon could be that patients with ITP have lower PLT counts than those with ET, and therefore, there are smaller denominators when calculating the coefficient of variation, resulting in a larger value. Further biological studies are required to validate these hypotheses.

We subdivided patients with ITP into 2 groups according to their PLT counts by using a PLT count cut-off value of 45 × 103 per μL, which is generally used to distinguish severe from moderate thrombocytopenia in South Korea. In most cases, MPV and PDW tended to be higher in patients with lower PLT counts (<45 × 103/μL) than in patients with higher counts (≥45 × 103/μL), despite that the difference was not statistically significant. This finding suggests that PLT destruction is increased in ITP, as demonstrated by the higher PDW, and that compensatory thrombopoiesis is more active, as demonstrated by the higher MPV. Again, this hypothesis ought to be tested in a future study.

In contrast, patients with ET showed lower PDW and MPV than healthy individuals. This finding contrasted with our expectation that MPV and PDW would be elevated in ET, owing to the increased production of PLTs, and suggested that clonal PLT production is a complex process. Regarding MPV and PDW in ET, there are few previous studies with which to compare our results. In an investigation comparing MPV and PDW in ET, reactive thrombocytosis, and healthy individuals, Sehayek et al[15] reported that ET was characterized by a 5-fold increase in small PLTs less than 7.5 fL and a 3-fold increase in larger-sized PLTs, resulting in lower MPVs in patients with ET than in those with reactive thrombocytosis, who in turn had lower MPVs than healthy controls.

The findings of Sehayek et al may explain our observations. However, a PDW of 10.5 or greater was reported in 50%, 21%, and 14% of patients with ET, patients with reactive thrombocytosis, and healthy controls, respectively, in their study. In contrast, Arellano-Rodrigo et al[12] reported that patients with ET had significantly higher MPV values than controls (n = 55 in each group). They also reported that the PDW was elevated in patients with ET, which they attributed to possible PLT activation and excessive heterogeneity in PLT volume in those patients.

Our observations may suggest that newly generated young PLTs in ET are not necessarily large as previously thought but vary in size. Alternately, our observations may hint at the speed at which PLTs decrease in size as they age. Although the production of young, large PLTs is increased, these PLTs quickly shrink and remain small for most of their life spans, resulting in a high proportion of small-sized circulating PLTs in ET. Because ET is a malignant, clonal process, PLTs generated in patients with this condition might not be the same as those produced in normal or compensatory conditions, in terms of size and lifespan.

Despite their statistical significance, the actual differences in MPV and PDW values between healthy individuals and patients with ET were fairly small (only 5.7% in MPV and 7.1% in PDW), in relation to the median values of healthy individuals (ie, when the difference was expressed as a percentage using the median values of healthy individuals as denominators). Therefore, our data may not be sufficiently conclusive; hence, a further analysis with a larger number of study subjects will be helpful to clarify and potentially validate our findings.

The analysis of MPV and PDW in patients with ET divided according to their PLT counts support our explanations, as presented earlier herein, as do the findings of Sehayek et al,[15] who report that MPVs were found to be significantly lower in patients with PLT counts of 770 or greater × 103 per μL than in those with PLT counts of less than 770 × 103 per μL. This finding suggests that the more advanced the disease, the higher the proportion of relatively small-sized circulating PLTs (as is the case in ET). This finding also suggests that the increased production of PLTs does not necessarily lead to a higher proportion of large PLTs but of small ones.

The PDW was also lower in patients with PLT counts of 770 or greater × 103 per μL than in those with counts less than 770 × 103 per μL, which suggests that the more advanced the ET, the more uniform the sizes of the PLTs, with a probable shift towards small-sized ones. Further studies on the role of age on PLT sizes in healthy individuals and in patients with ET would help address these theories.

One limitation of our study was that the distributions of age and sex in the healthy individual, ITP, and ET groups were different. However, in a study of 306 individuals (101 male and 205 female), Giovanetti et al[16] showed that MPV and PDW do not differ significantly with respect to sex and age except when comparing those values in the 1- to 10-year-old age group to those in most other age groups (the 10–20-year-old age group was the exception). Our analysis also showed that neither the MPV nor the PDW correlates with age and sex in healthy individuals.[2]

Another limitation of our study was that data on immature platelet fractions (IPFs) were not included. The IPF among circulating PLTs, which can be measured using automated hematologic analyzers, directly reflects the status of thrombopoiesis from the bone marrow. The results from previous studies[6,8,17,18] showed that the IPF could be a useful marker for distinguishing ITP from other types of thrombocytopenia, including acute lymphoblastic leukemia, thrombocytopenia due to chemotherapy, bone-marrow failure, or hereditary macrothrombocytopenia. Further, IPF predicts PLT recovery and can thus serve as a gauge for requiring PLT transfusion in patients who undergo hematopoietic stem-cell transplantation or chemotherapy.[19,20] Studies of IPF in patients with ET are relatively rare; Kissova et al[21] showed that IPF positively correlated with JAK2 V617F mutation status and thrombotic complications.

Because our study was performed retrospectively in patients lacking IPF data, we were not able to assess their IPF status, despite that doing so would have provided useful information. Nevertheless, our study has the advantage of having analyzed MPV and PDW in the context of understanding PLT biology in ITP and ET—these parameters reflect the overall features of immature and mature circulating PLTs. Further, both these parameters indirectly reflect IPF to some extent. A future study that includes IPF data will be helpful in potentially elucidating PLT biology in terms of the pathophysiologies of ITP and ET.

In conclusion, we discovered alterations in the composition of the circulating PLTs of patients with ITP and those with ET using MPV and PDW values; to our knowledge, this study (in which we also compared healthy individuals) is the first of its kind. MPV and PDW were increased in ITP, suggesting an upregulation in the production of young, large PLTs in conjunction with the indiscriminate destruction of circulating PLTs regardless of their age. Separately, MPV and PDW were decreased in patients with ET, suggesting that the proportion of small PLTs was higher than normal although PLT production is increased. The more severe the disease, the stronger the tendency for the MPV and/or PDW to be altered—this principle is consistent with our interpretations herein.