Article Text

Download PDFPDF
The prognostic value of mean platelet volume in decompensated heart failure
  1. Hayati Kandis1,
  2. Hakan Ozhan2,
  3. Serkan Ordu2,
  4. Ismail Erden2,
  5. Onur Caglar2,
  6. Cengiz Basar2,
  7. Subhan Yalcin2,
  8. Recai Alemdar2,
  9. Mesut Aydin2
  1. 1Duzce University, Düzce Medical School, Department of Emergency Medicine, Konuralp Düzce, Turkey
  2. 2Duzce University, Düzce Medical School, Department of Cardiology, Konuralp Düzce, Turkey
  1. Correspondence to Dr Hakan Ozhan, Duzce Universitesi, Duzce Tip Fakültesi 81620, Konuralp Duzce, Turkey; ozhanhakan{at}


Background Congestive heart failure (CHF) is a major public health problem that is related to substantial morbidity, impaired quality of life and diminished survival. Mean platelet volume (MPV) is an indicator of platelet activation.

Aim To investigate whether there is a difference of MPV in patients with decompensated and stable heart failure (SHF), and test the prognostic value of MPV in decompensated heart failure (DHF).

Methods 136 consecutive patients with DHF were enrolled. 71 with SHF were also enrolled for comparison. Patients were followed up for a mean of 18±12 months. The primary endpoint was death from any cause. Clinical characteristics of patients with DHF who died during follow-up were compared with the those of the survivors.

Results MPV was significantly higher in DHF group than in the SHF group. 71 patients died during the follow-up period (18±12 months). Comparison with survivors revealed that mortality was associated with age, systolic blood pressure, pulmonary artery pressure, serum creatinine, urea and MPV. MPV was determined as an independent risk factor for mortality (OR 1.553, 95% CI 1.024 to 2.354, p=0.038). Receiver operating characteristic analysis showed that MPV level on admission was a predictor of mortality (area under the curve (AUC) for in-hospital mortality was 0.716 (95% CI 0.632 to 0.789, p=0.003) and AUC for 6-month mortality was 0.815 (95% CI 0.74 to 0.877, p<0.001), respectively).

Conclusion MPV is increased in patients with DHF. Also, MPV on admission is an independent predictor of in-hospital mortality and 6-month mortality.

  • Heart failure
  • mean platelet volume
  • cardiac care

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Congestive heart failure (CHF) is a major public health problem that is related to substantial morbidity, impaired quality of life and diminished survival.1 Acute decompensation in patients with chronic heart failure is among the most common indications for hospitalisation.2 Patients hospitalised with decompensated heart failure (DHF) are at high short-term risk of mortality and morbidity, with short-term readmission rates ranging from 15% to 50%.3 CHF is characterised by increased thromboembolic events, stroke and sudden death. This may be related to a prothrombotic state, abnormal thrombogenesis, endothelial dysfunction and platelet abnormalities.4 An abnormal state of platelet activation may contribute to the pro-coagulant state in congestive heart failure.5 The exact role of platelets in the pathogenesis of acute decompensated heart failure is unclear.

Mean platelet volume (MPV) is an indicator of platelet activation and there is strong evidence indicating than MPV is an important variable and that larger platelets have a higher thrombotic potential.6 Larger platelets are more dense, aggregate more rapidly on collagen challenge, produce greater amounts of thromboxane B2, release more serotonin and β-thromboglobulin, and express more Ib and IIb/IIIa glycoproteins than smaller platelets.6 Elevated MPV has been recognised as an independent risk factor for myocardial infarction and stroke.7 8

The prognostic value of MPV in patients with decompensated heart failure has not been studied extensively. Our primary aim in the present study is to investigate whether there is a difference of platelet indices in patients with decompensated and stable heart failure. We have also tested the prognostic value of platelet indices in the decompensated heart failure patients.



This retrospective study was conducted in Duzce University Medicine Faculty Hospital. A total of 136 consecutive patients (77 men and 59 women) who had been admitted to the Duzce University emergency clinic with decompensated heart failure between December 2006 and May 2009, and who had fulfilled the inclusion criteria, were enrolled. Seventy-one (50 men, 21 women) patients with stable heart failure who had not been hospitalised with DHF during the last 6 months were also enrolled for comparison. The emergency department is the city's largest clinic and receives ambulance cases. The emergency fellow (HK) had identified the DHF patients in the study and consulted with cardiology department. Two cardiologists (SO and IE) had performed transthoracic echocardiograms before hospitalisation to the cardiology department, which was the routine procedure in our clinic before admission of heart failure cases. They also had performed echocardiographic measurements of stable heart failure patients. After establishing the hypothesis (HO) a datasheet was prepared. A follow-up time of at least 6 months was proposed. Patients were collected according to ICD codes at baseline; those who fulfilled the inclusion criteria were enrolled. DHF was defined as acute change in the clinical status of the patient, from a baseline NYHA of 1–3 to NYHA class 4 and admission of the patient to the emergency clinic. Stable heart failure was defined as NYHA class 1–3 patients with an ejection fraction of <45%, who had been followed up at the outpatient cardiology clinic and NYHA class was unchanged in the last 6 months. The emergency department fellow was responsible for the recruitment of acute heart failure patients, whereas five cardiologists were involved in the recruitment of stable heart failure patients (CB, MA, RA, SY, OC). The recruitment was done consecutively. Exclusion criteria were recent (<3 months) acute coronary syndrome, myocardial infarction, cerebrovascular accident and a change of NYHA class during the last 6 months. Subjects with acute or chronic liver or renal disease, active hepatitis or malignancy, and patients who used of antiplatelet medication other than aspirin were also excluded.

Demographic information, medical history, clinical characteristics and laboratory test results were collected from hospital medical records. Bloods were sampled before recruitment, at the time of admission. Complete blood count, basic biochemical measurements and transthoracic echocardiographic measurements could be collected in all patients since all of these tests are routine procedures in DHF in our hospital. All the investigators were blinded to results of blood tests prior to selection of patients. Follow-up information was obtained by the review of medical records at all participating centres and contact with the patient or his/her family. The Institutional Review Board of the Duzce University Faculty of Medicine approved the study, a waiver of consent was granted and patient identity was protected. The local ethics committee has also approved the study.

Follow-up and endpoints

Patients were retrospectively followed up (mean follow-up duration was 18±12 months). The primary endpoint was death from any cause. Clinical characteristics of DHF patients who died during follow-up were compared with those of the survivors.

Laboratory tests

A 10 ml aliquot of fasting blood was drawn from each subject on admission and the samples were immediately centrifuged. Serum glucose levels were measured with a standard hexokinase reference method; serum triglyceride, high density lipoprotein, urea nitrogen, creatinine (Jaffe method without deproteinisation) and uric acid levels were determined by standard clinical chemistry methods using an autoanalyser, Architect C8000 (Abbott Diagnostics, Tokyo, Japan).

Complete blood count samples which were drawn into vacutainer tubes containing 0.04 ml of the 7.5% K3 salt of EDTA were analysed within 1 h after sampling with a commercially available analyser (Sysmex XT 2000i, Roche Diagnostics, Tokyo, Japan).

Echocardiographic measurements

The examinations were performed by an experienced sonographer (SO or IE) who was blinded to patient data on the Vivid 3 echocardiography machine (General Electric, Horten, Norway) using 2.5 MHz transducers. Two-dimensional echocardiographic calculations were obtained by parasternal long axis, apical three- and four-chamber views. Left ventricular ejection fraction was calculated by the Teicholz formula. Colour flow imaging was performed from the left parasternal and apical long-axis views.

Statistical analyses

All values are given as mean±SD. SPSS V.15 was used for analysis. The unpaired Student t test was used for group comparisons. Categorical data were compared with the χ2 test. Multiple group comparisons were performed with analysis of variance. The predictive accuracy of the model was graphically displayed by the receiver operating characteristic (ROC) curves (Medcalc software; Medcalc, Mariakerke, Belgium). The ROC curve is a plot of sensitivity versus 1-specificity for different threshold probabilities of mortality. A p value of <0.05 was considered significant.


Table 1 compares the clinical characteristics of the subjects with DHF and patients with stable heart failure. The frequency of hypertension, diabetes mellitus, coronary artery disease, atrial fibrillation and drug use was similar in the decompensated and stable heart failure groups. Echocardiographic variables, fasting glucose, total cholesterol and platelet count were similar between the two groups. However, mean platelet volume was significantly higher in the DHF compared to the stable heart failure group.

Table 1

Comparison of clinical characteristics of the patients with decompensated heart failure and stable heart failure

Of the 136 DHF patients, 17 died during hospital follow-up. Twenty-seven patients died in the first 6 months of follow-up after discharge from the hospital. A total of 71 patients (52%) died during the follow-up period (18±12 months). Table 2 compares the clinical variables of survivors and patients who died. The patients who died during follow-up were significantly older than the survivors and had higher mean pulmonary artery pressure, urea, systolic blood pressure, serum creatinine and mean platelet volume.

Table 2

Comparison of clinical characteristics of patients with decompensated heart failure who died during follow-up and those who survived

In order to understand whether MPV is an independent determinant for increased mortality, logistic regression analysis was performed. Age, MPV, urea, systolic blood pressure on admission, haemoglobin and sodium were used as covariates. MPV was found to be the independent predictor of mortality (OR 1.553, 95% CI 1.024 to 2.354, p=0.038) (table 3).

Table 3

Independent predictors of mortality by logistic regression analysis

ROC analysis showed that MPV levels >10.5 fl on admission could predict in-hospital mortality with 76% sensitivity (95% CI 50% to 93%) and 54% specificity (95% CI 44% to 62%), and 6-month mortality with 82% sensitivity (95% CI 68% to 92%) and 66% specificity (95% CI 57% to 75%). Area under the curve (AUC) for in-hospital mortality was 0.716 (95% CI 0.632 to 0.789; p=0.003) and AUC for 6-month mortality was 0.815 (95% CI 0.74 to 0.877, p<0.001) (figure 1A,B). Using this cut-off value, Kaplan–Meier analysis showed that mortality rate was significantly higher in patients who had higher MPV levels (p=0.003) (figure 2).

Figure 1

Receiving operator characteristic curves for mean platelet volume for predicting in-hospital and 6-month mortality rates.

Figure 2

Kaplan–Meier curve for all-cause mortality according to mean platelet volume (MPV). The mortality was significantly higher in patients with high (>10.5 fl) MPV (64.3% vs 39.0%; p=0.003).


The present study showed that MPV increases in patients with DHF. Furthermore, MPV is an independent predictor of mortality at follow-up. The best cut-off value showing increased risk is >10.5 fl.

Several reports have been published in the literature that described abnormal platelet function and size in CHF.4 5 Chung et al reported that acute heart failure patients showed abnormal platelet activation (higher platelet surface P selectin) when compared to stable CHF patients and healthy controls. They have also shown that the difference of MPV was insignificant in stable and acute heart failure patients. Contrary to this finding, we have shown with a much larger cohort (22 vs 136) that MPV was higher in patients with DHF compared to stable heart failure patients.5 All myocardial infarction patients with severe cardiac failure had larger platelet volumes than patients with mild or no failure. In the present study we have further shown that increased MPV is a prognostic variable in heart failure. The findings of Hendra et al were similar; they showed that myocardial infarction patients with severe cardiac failure had larger platelet volumes than patients with mild or no failure.9

Congestive heart failure is characterised by increased thromboembolic events, stroke and sudden death. Dilated cardiac chambers, poor contractility, regional wall abnormalities and concomitant atrial fibrillation may all predispose to thromboembolism by facilitating stasis of intracardiac blood flow.10 The cause of platelet activation/dysfunction is not clear. Jafri et al reported that patients with severe heart failure with high plasma norepinephrine concentration or low ejection fraction were more likely to have activation of platelets (higher plasma platelet factor 4, β-thromboglobulin) and the coagulation system.11 There was a significant relationship between platelet activation/abnormalities, sympathetic nervous systems overactivity and poor prognosis of heart failure.11 Taking into account that MPV is a well known strong indicator of platelet activation, our findings about MPV in acute decompensated heart failure may be related to the abnormal activation of platelets during decompensation.

Increased MPV levels were shown in patients with acute myocardial infarction and unstable angina.12–14 Huezek et al reported that measurment of MPV helps to select a subgroup of patients (MPV >10.3 fl) with significantly higher mortality in 6-month follow-up in ST elevation myocardial infraction.7 Moreover, Martin et al have shown that higher MPV, when measured 6 months after myocardial infarction, is associated with increased risk of death and reinfarction during a 2-year follow-up.14 A similar association of MPV in DHF is not unacceptable. Abnormalities of platelet function may contribute to the relatively poor prognosis in DHF patients. To our knowledge, the present study is the first to display a significant prognostic value of increased MPV in patients with DHF. MPV levels >10.5 fl in particular indicate a poor prognosis.

Clinical implications of the current results are limited and the results should be interpreted with caution. In our patient cohort, patients who died during follow-up had higher levels of urea and serum creatinine, and higher systolic blood pressure, similar to the results of the ADHERE study,15 which showed that a patient's initial systolic blood pressure and levels of BUN and creatinine are especially important in long-term prognosis; the prognostic efficacy of platelet indices has not been studied in large registries. Therefore, our findings should be confirmed with multicentre prospective studies.

Limitations of the study

The results cannot be applied to diastolic heart failure since all the patients had systolic dysfunction. The study contained a fairly low number of subjects. Laboratories have different levels of normal ranges for MPV so different cut-off values using local laboratory data may be necessary. The overall increase in mortality carries a limited potential of clinical significance, and further prospective studies with larger cohorts are mandatory to maintain results that will have prognostic utility to emergency physicians.



  • Competing interests None.

  • Ethics approval This study was conducted with the approval of the Duzce Medical school Ethics Committee.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Linked Articles

  • Primary survey
    Kevin Mackway Jones