Platelet Size Predicts Mortality in Patients With Sepsis

Daniel M. Keller, PhD

July 10, 2012

July 10, 2012 (Bangkok, Thailand) — The size of platelets can predict the risk for death in patients with sepsis. Preliminary data indicate that mean platelet volume (MPV) is an easily accessible prognostic marker of mortality in sepsis.

Thomas Grünewald, MD, PhD, assistant professor of internal medicine and head of the division of infectious diseases and tropical medicine at the Klinikum St. Georg in Leipzig, Germany, who presented a poster here at the 15th International Congress on Infectious Diseases, said that this is one of the few reports on MPV as a biomarker in sepsis, and it strongly supports its usefulness.

An elevated level of procalcitonin is another biomarker that supports the clinical diagnosis of sepsis, but that is relatively expensive to do. In addition, scoring systems such as the Acute Physiology and Chronic Health Evaluation (APACHE) II/III and the Sepsis-related Organ Failure Assessment (SOFA) can be used as prognostic indicators.

"What is lacking up to now is a biomarker for outcome — who gets worse outcome, who gets a good outcome, who will survive, who will not," Dr. Grünewald told Medscape Medical News. "We saw in veterinary journals a few reports of mean platelet volume [as] a marker of outcome in sepsis in...cats and cattle and dogs. So we thought, why don't we try it in humans?"

In this study of 191 patients with documented sepsis in his institution, Dr. Grünewald and colleagues prospectively evaluated MPV — at admission, at the onset of sepsis, at the diagnosis of sepsis, and during the course of the disease — as a marker for the prediction of outcomes. These data were compared with data from patients with acute upper and lower gastrointestinal (GI) bleeding, who served as control subjects. Data on other standard laboratory biomarkers and clinical parameters were collected at the same time.

For patients with sepsis, median age was 72 years (range, 21 to 97 years), and 38.2% were women. For control subjects (n = 56), median age was 74 years (range, 29 to 95 years), and 45.5% were women.

Sepsis was most often caused by Staphylococcus aureus (27.7% of cases) or by Escherichia coli (22.0%). The most common sites of infection were the urinary tract, skin and soft tissues, and the blood stream, mostly related to catheter use.

On admission, there was no difference in MPV between the sepsis and control groups (median, 9.10 vs 9.10 fL). The median level of procalcitonin was higher in the sepsis group than in the control group (2.07 ± 78.13 mg/L vs 0.16 ± 13.12 mg/L), but in both cases the standard deviations were large, indicating widely varying levels. Median platelet numbers were similar in the sepsis and control groups (208.5/nL vs 253.5/nL).

In the sepsis group, 41 patients (21.5%) died. MPV on admission and at the onset of symptoms was strongly associated with death. The researchers tested various receiver-operator curves and determined that the best predictor of death or survival was when they used an MPV discriminatory value of 8.7 fL. This figure is at the high end of the normal range of platelet volumes, but is still highly predictive of poor outcome, namely death, in patients with sepsis.

"We saw that patients with MPVs higher than 8.7 are much more prone — the odds ratio is 3.2 — to worse outcomes (for example, death) than patients with MPVs lower than 8.7," Dr. Grünewald said. "If you measure MPV at admission, the P value is .001; if you measure it at the beginning of symptoms (like fever or shivering or something like this), the P value is .0005. It's highly significant."

An analysis of 183 of the patients showed that those who died had a higher MPV than survivors (9.6 vs 9.19 fL; P = .031). At the time of positive blood cultures, those values had increased (11.2 vs 9.7 fL; P = .008). Parameters of temperature, leukocyte count, lactate, procalcitonin, and C-reactive protein values were not at all predictive of outcome (P > .1 for all).

In contrast, MPV was not predictive of outcome in the control group of patients with GI bleeding, another population with activated platelets. "In patients with high MPV values and GI bleeding, there was no correlation with outcome.... It's just for sepsis. We think the reason for this is the endothelial activation, which is the most important factor for the size of the platelets," Dr. Grünewald explained.

His team plans to do a retrospective chart review of all patients with sepsis at his institution to evaluate the reliability of MPV as a marker for poor outcome, and then plans to do a large prospective study.

He explained that if he sees a sepsis patient with elevated MPV, he might attempt more aggressive therapy.

Dr. Grünewald noted that MPV is easily obtained because it "is like a waste product of a complete blood count. You will always get [the MPV] if you have a cell sorter...but nobody noticed before."

Daniel De Vos, PhD, research manager and senior scientist at the Burn Wound Center, Military Hospital Queen Astrid, in Brussels, Belgium, who was not involved in the study, told Medscape Medical News that "it's quite a simple system.... It's better than the other markers — procalcitonin and [C-reactive protein] — because they are still controversial and sometimes they work and sometimes [they don't]." MPV might be more specific, and is quite cheap because the cell sorter already produces an MPV value, he said. "If it's validated, I'd surely take notice of it."

There was no commercial funding for the study. Dr. Grünewald and Dr. De Vos have disclosed no relevant financial relationships.

15th International Congress on Infectious Diseases (ICID): Abstract 45.021. Presented June 15, 2012.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.