Home About us Editorial board Ahead of print Current issue Archives Submit article Guidelines Contacts Login 
ISSN: Print -2349-0977, Online - 2349-4387

 Table of Contents  
Year : 2018  |  Volume : 4  |  Issue : 4  |  Page : 205-209

Evaluation of platelet indices as additional diagnostic tool for neonatal sepsis

1 Department of Pediatrics, All India Institute of Medical Sciences, Jodhpur, Rajasthan, India
2 Department of Pediatrics, Vardhman Mahavir Medical College and Safdarjung Hospital, New Delhi, India
3 Department of Pediatric Gastroenterology, Medanta – The Medicity, Gurgaon, India
4 Department of Hematology, Vardhman Mahavir Medical College and Safdarjung Hospital, New Delhi, India

Date of Web Publication29-Oct-2018

Correspondence Address:
Sugandha Arya
Department of Pediatrics, Vardhman Mahavir Medical College and Safdarjung Hospital, New Delhi
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/astrocyte.astrocyte_8_18

Rights and Permissions

Introduction: Search for newer markers to enhance sensitivity and specificity of the existing sepsis screen and attempts toward this has been in place for a long time. Platelet indices are one such marker. Materials and Methods: Babies with signs and symptoms or born with risk factors for sepsis were enrolled. Those with positive culture or with clinical sepsis as per Centre for Disease Control definition were classified under the group “cases” (n = 188), whereas all neonates initially suspected to have sepsis but who had a negative blood culture and no clinical sepsis as per definition were classified under the group “control” (n = 188). Blood culture, sepsis screen, and platelet indices [platelet count, mean platelet volume (MPV), platelet distribution width (PDW)] were performed on all these babies. Results: The platelet count was decreased, whereas PDW and MPV were increased in septic babies (P < 0.0001). Thrombocytopenia and increased MPV were frequently observed in babies with late-onset sepsis (P = 0.012). Thrombocytopenia was the most predictive marker for culture positivity in septic babies (83.08%), and when all the platelet indices (MPV + PDW + PC) or (MPV + PDW) were combined (46.34%) it was found to be highly specific marker for prediction of sepsis. Platelet indices had a better sensitivity (83.08%) than sepsis screen (60%). When sepsis screen and platelet indices were combined, it increased the specificity (62.6%). The receiver operating characteristics curve suggested that MPV is a good marker having highest area under curve. Conclusion: Although data on platelet indices are still nascent, platelet indices may be used as a sensitive marker and combined with sepsis screen to exclude a non-septic case.

Keywords: Mean platelet volume, neonatal sepsis, platelet distribution width, platelet indices, thrombocytopenia

How to cite this article:
Mittal A, Arya S, Charan LS, Saluja S, Chellani H. Evaluation of platelet indices as additional diagnostic tool for neonatal sepsis. Astrocyte 2018;4:205-9

How to cite this URL:
Mittal A, Arya S, Charan LS, Saluja S, Chellani H. Evaluation of platelet indices as additional diagnostic tool for neonatal sepsis. Astrocyte [serial online] 2018 [cited 2022 May 29];4:205-9. Available from: http://www.astrocyte.in/text.asp?2018/4/4/205/244300

  Introduction Top

Neonatal sepsis is an important cause of neonatal morbidity and mortality worldwide. However, it is a diagnostic challenge as there are overlapping signs and symptoms which preclude a specific diagnosis of sepsis. So, we have to rely on investigations to guide us. Blood culture has always been the gold standard for the diagnosis of neonatal sepsis. It has been noted that only 20% of symptomatic neonates with suspected early-onset sepsis (EOS) have a positive blood culture, and only 30% neonates clinically suspected to have late-onset sepsis (LOS) in neonatal intensive care unit (NICU) setting have a positive blood culture.[1],[2] However, the blood culture report is available too late and it cannot be relied upon for making immediate decisions.

To overcome these limitations, we usually rely on sepsis screening. But it has a variable sensitivity and specificity. The negative predictive value of these parameters is too low to confidently rule out sepsis.[3],[4] The limitations of blood culture, its low positivity rates, and poor diagnostic capability of sepsis screen in neonates make the diagnosis of sepsis difficult, and thus the need for better diagnostic parameters arises.

There have been studies showing significant changes in platelet indices in patients with neonatal sepsis.[5],[6],[7],[8],[9],[10] These studies have measured platelet count, mean platelet volume (MPV), and platelet distribution width (PDW). It has been shown by these studies that platelet count decreases and MPV and PDW increases in neonates with sepsis.[6]

Despite being a promising and convenient marker, there have been only a few studies on their utility in neonatal sepsis. Most of the studies that have compared platelet indices in newborns have focused on culture positive sepsis only and those with culture-negative sepsis have not been included. None of the available data compare platelet indices with the existing sepsis screen in prediction of neonatal septicemia.

This study is an effort to fulfil the gap in the existing literature and to determine whether this could act as a tool to enhance our ability to diagnose neonatal sepsis early. We conducted this study to evaluate platelet indices (platelet count, PDW, MPV) as marker of neonatal sepsis and to determine whether there is a difference in platelet indices among term versus preterm, EOS versus LOS, and Gram-positive versus Gram-negative sepsis.

  Materials and Methods Top

This study was a prospective case–control study, conducted in the neonatal division of Department of Paediatrics and the Department of Haematology and Microbiology of Vardhman Mahavir Medical College and Safdarjung Hospital, New Delhi, over a period of 1 year. Clearance was obtained from institutional ethics committee.

Neonates with culture-positive sepsis or clinical sepsis as per Centre for Disease Control (CDC) definition were categorized as cases and neonates initially suspected of having sepsis but with negative blood culture and not fulfilling the criteria of clinical sepsis as per CDC definition were categorized into the control group. Babies with congenital and acquired causes of thrombocytopenia other than sepsis (e.g. autoimmune disorders of platelets, allo-immune disorders of platelets) were excluded from the study.

All babies enrolled were investigated for blood culture, sepsis screen [C-reactive protein (CRP), total leukocyte count (TLC), absolute neutrophil count (ANC), immature to total neutrophil (IT) ratio], and platelet indices (platelet count, MPV, PDW). For this, approximately 2 mL of venous blood was drawn from each neonate through peripheral veins.

Based on the reports of blood culture and sepsis screen, the neonates were classified into culture-positive sepsis, clinical sepsis with negative blood culture, and no sepsis on the basis of definitions provided by the CDC.[11]

The data were analyzed using SPSS version 11.0. The mean and median values were used as applicable. Paired t-test and Fisher's exact test were used where applicable. Receiver operating characteristics (ROC) curve was prepared, and conclusions were drawn.

  Results Top

A total of 376 babies were enrolled, of which 188 neonates were categorized as group “cases” (clinical sepsis 123, culture-positive sepsis 65) and those (188) neonates without sepsis were categorized as group “controls.” The demographic profile of the two groups was comparable. There was a larger number of males, but platelet indices did not differ significantly with gender, so this difference in gender distribution did not confound our result [Table 1].
Table 1: Distribution of Patients by Age, Sex, Weight, and Gestational Age in Case and Control Groups

Click here to view

The babies in “cases” group had a greater number of subjects with higher (than defined cut-off)[5] platelet indices than the control group (P < 0.05) [Table 2].
Table 2: Percentage of Positive Subjects Above Platelet Indices' Cut-Off

Click here to view

The mean platelet count was lower in “cases” group (1.098 ± 0.747 lakhs/mm3) than in the control group (2.038 ± 0.762 lakhs/mm3; P < 0.0001). The mean MPV was higher in the case group (11.85 ± 1.716 fl) than in the control group (9.81 ± 1.460 fl), P < 0.0001. In addition, the mean PDW value was higher in the case group (20.68 ± 2.239) than in the control group (18.69 ± 1.974), P < 0.0001.

The sensitivity and specificity of platelet indices for diagnosis of neonatal sepsis were assessed by comparing them against blood culture which is the gold standard for diagnosis of neonatal sepsis [Table 3].
Table 3: Performance Variables of Platelet Indices for Diagnosis of Neonatal Sepsis with Blood Culture Being Gold Standard (n=188)

Click here to view

It was seen that thrombocytopenia was the most sensitive marker (83.08%) followed by MPV and PDW in detecting babies with culture-positive sepsis. However, it has a low specificity (20.33%). But when we combine MPV and PDW or combined all the three markers (MPV + PDW + PC), the specificity increased to 46.34%.

The area under curve (AUC) is 0.156 for platelet count; it is 0.814 for MPV and 0.743 for PDW on ROC curves [Figure 1].
Figure 1: ROC Curves for Correlation of Platelet Indices- Platelet Count, MPV, PDW with Gold Standard, i.e., Blood Culture.

Click here to view

The sensitivity of sepsis screen in this study was 60% (which was lower than the sensitivity of platelet indices) and specificity was 31.71%. However, when we combine sepsis screen and platelet indices, the specificity for diagnosis of neonatal sepsis increased to 62.6% in our group of babies [Table 4].
Table 4: Correlation Between Markers of Platelet Indices with Positive Sepsis Screen and Gold Standard Blood Culture (n=188)

Click here to view

There was no significant difference in platelet indices among the preterm and term babies with sepsis.

When comparing babies with EOS and LOS, it was found that there were more patients with thrombocytopenia in the LOS group (87.38%) than in the EOS group (72.94%) (P = 0.012). In addition, more babies with LOS had higher MPV (78.64%) than in the EOS group (61.18%) (P = 0.009). Again, patients with increased PDW were 72.82% in LOS group, whereas they were 60% in EOS group (P = 0.063).

There were 65 patients who had Gram-positive or Gram-negative growth on blood culture. While comparing the prevalence of thrombocytopenia in Gram-positive and Gram-negative sepsis, it was found to be 90.48% and 79.55%, respectively. Whereas increased MPV was seen in 76.19% and 79.55% babies with Gram-positive and Gram-negative growth on culture, respectively (P = 0.758).

  Discussion Top

We have seen that the groups differed in terms of prevalence of thrombocytopenia and increased MPV and PDW, which was statistically higher (P < 0.0001) in the cases when compared with controls. In the study group, 80.9% of cases of neonatal sepsis developed thrombocytopenia.

The prevalence of thrombocytopenia in sepsis is variable and different values have been reported by workers across the globe. Abdulla et al. in their study showed that 42.8% babies with sepsis developed thrombocytopenia.[12] In another study by Ahmad et al., it was seen that mortality rate was higher among children with thrombocytopenia and its prevalence was 24.7% in neonatal sepsis.[13] This variation may be attributed to the fact that in this study we considered a cut-off of platelet counts of less than 1.5 lakhs/mm3 and other contemporary studies considered a cut-off of 1 lakhs/mm3. The sensitivity and specificity of thrombocytopenia in detecting neonatal sepsis were found to be 83.08% and 20.33%, respectively, when compared with other studies where sensitivity was 24.7%.[13] Many other investigators have demonstrated thrombocytopenia as an important marker of neonatal sepsis which this study reinforces.[12]

MPV in our data was increased in 70.7% of neonatal sepsis which was far greater when compared with the results of the study by Abdulla et al. where 27.8% children suffering from sepsis had an increase in MPV.[12] It was also seen that sensitivity and specificity of increase in MPV in diagnosis of neonatal sepsis were found to be 78.4% and 33.3%, respectively. In a study by Arad et al., it was seen that sensitivity and specificity of an increase in MPV in diagnosis of neonatal sepsis were 54% and 46%, respectively.[8]

PDW was found to be increased in 67% cases (n = 188) when compared with controls (36.2%) (P < 0.0001). An increase in PDW was far greater when compared with the results of the study by Abdulla et al. which show that PDW increased in 38% cases of sepsis.

On comparing the platelet indices with the gold standard, the blood culture, thrombocytopenia (platelet count < 1.5 lakhs/mm3) was the most sensitive marker for sepsis, (83.08%). Previous studies have shown the sepsis screen to have variable sensitivity and specificity, because the timing of sample collection may affect its results. If sample is drawn several hours after illness or after starting antibiotics, sensitivity and specificity of sepsis screen will change. Moreover, CRP increases in other inflammatory conditions as well.[14]

When we combine the existing sepsis screen and platelet indices and correlate it with blood culture, the specificity of diagnosis increases (46%) [Table 4]. A high specificity was seen with a combination of sepsis screen with MPV + PDW + platelet count or sepsis screen with MPV + PDW which was 62.6% in both cases. However, there are no published data on correlation of sepsis screen with platelet indices for evaluation for neonatal sepsis. But the data from this study seem to indicate that a combination of sepsis screen with platelet indices can be used to reasonably rule out sepsis in each scenario.

As AUC was highest for MPV, it was concluded that it was the most promising marker. There were no significant differences in MPV, PDW, and platelet counts (P > 0.05) on comparing preterm and term babies. Some of the investigators such as Patrick and Lazarchick have reported a higher incidence of thrombocytopenia in preterm.[5] However, there is also literature that shows no difference in platelet indices in septic babies whether they are full-term or not.[12]

When we compared platelet indices in babies with EOS and LOS, it was seen that prevalence of thrombocytopenia was significantly higher in LOS which is supported from a study by Kudawla et al. who also concluded that decreased platelet count was seen more frequently in LOS group (P < 0.05).[15] We found a significantly higher MPV in the LOS group and the finding is also well-supported by other studies.[5],[15]

However, the difference in PDW was not significant in this study, whereas other similar studies in literature have found it to be significantly high in LOS babies.[16]

In this study, a larger number of babies with Gram-positive sepsis had thrombocytopenia (platelet count < 1.5 lakhs/mm3) and increased MPV, whereas babies with Gram-negative infection had increased PDW more frequently. However, these differences were not statistically significant. The finding is aptly supported in previous studies, where it has been noted that no difference in thrombocytopenia was noted in babies with neonatal sepsis due to Gram-negative or Gram-positive organisms (P = 0.17). Absence of significant difference in increased MPV was also supported by other studies from literature.[12]

Although this study did not report any significant difference in PDW, Patrick and Lazarchick have shown in their study that PDW increased in 44.35% of Gram-positive and 65.5% in Gram-negative sepsis which means significantly high PDW was seen in Gram-negative sepsis.[5]

  Conclusions Top

It was concluded from this study that platelet indices may serve as an important tool to aid sepsis screening. The platelet count decreased with development of sepsis and PDW and MPV increased in septic babies. Platelet indices did not differ significantly with gestational age nor with Gram-positive or Gram-negative blood culture. Thrombocytopenia was the most sensitive marker for culture-positive sepsis, and the highest specificity of platelet indices was seen when all the platelet indices (MPV + PDW + PC) or (MPV + PDW) were combined. Platelet indices alone seem to have a better sensitivity (83.08%) in identification of sepsis than sepsis screen (60%).

Thrombocytopenia and an increase in MPV were seen significantly more in babies with LOS; however, the increase in PDW was not significant.

When we combine sepsis screen and platelet indices, the specificity of diagnosis of sepsis increased to 62.6%. Thus, it may be concluded that platelet indices are sensitive markers to identify septic babies and they may be combined with the existing sepsis screen to specifically exclude the diagnosis of neonatal sepsis.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.

  References Top

Okascharoen C, Sirinavin S, Thakkinstian A, Kitayaporn D, Supapanachart S. A bedside prediction — Scoring model for late onset neonatal sepsis. J Perinatol 2005;25:778-83.  Back to cited text no. 1
Singh SA, Dutta S, Narang A. Predictive clinical scores for diagnosis of late onset neonatal septicaemia. J Trop Pediatr 2003;49: 235-9.  Back to cited text no. 2
Philip AG, Hewitt JR. Early diagnosis of neonatal sepsis. Pediatrics 1980;65:1036-41.  Back to cited text no. 3
Gerdes JS, Polin RA. Sepsis screen in neonates with evaluation of plasma fibronectin. Pediatr Infect Dic J 1987;6:443-6.  Back to cited text no. 4
Patrick RH, Lazarchick J. Effect of bacteraemia on an automated platelet measurement in neonates. Am J Clin Pathol 1990;93:391.  Back to cited text no. 5
Castle V, Andrew M, Kelton J, Giron D, Johnston M, Carter C. Frequency and mechanism of neonatal thrombocytopenia. J Pediatr 1986;108:749-55.  Back to cited text no. 6
Tate DG, Carlton GT, Johnson, Sorenson RL, Nesbit M, White J et al. Immune thrombocytopenia in severe neonatal infection. J Pediatr 1981;98:449-53.  Back to cited text no. 7
Arad ID, Alphan G, Sznajderman SD, Eldor A. The mean platelet volume (MPV) in the neonatal period. Am J Perinatol 1986;3:1-3.  Back to cited text no. 8
Moro ML, De Toni A, Stolfi I, Carrieri MP, Braga M, Zunin C. Risk factors for nosocomial sepsis in newborn intensive and intermediate care units. Eur J Pediatr 1996;155:315-32.  Back to cited text no. 9
Escobar GJ. The neonatal “sepsis work-up” personal reflections on the development of an evidence-based approach toward newborn infections in a managed care organization. Pediatrics 1999;103:360-73.  Back to cited text no. 10
Teresa CH, Mary A, Dudeck MA. CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infections in the acute care settings. Am J Infect Control 2008;36:309-32.  Back to cited text no. 11
Abdulla A, Maghayreh M, Khriesat W, Swedan S. The effect of neonatal sepsis on platelet count and their indices. Jordan Med J 2008;42:82-6.  Back to cited text no. 12
Ahmad MS, Waheed A. Platelet counts, MPV and PDW in culture proven and probable neonatal sepsis and association of platelet counts with mortality rate. J Coll Physicians Surg Pak 2014;24:340-4.  Back to cited text no. 13
Krediet T, Gerards L, Fleer A. The predictive value of CRP and I/T ratio in neonatal sepsis. J Perinat Med 1992;20:479-88.  Back to cited text no. 14
Kudawla M, Dutta S, Narang A. Validation of a clinical score for the diagnosis of late onset neonatal septicaemia in babies weighing 1000-2500 g. J Trop Pediatr 2008;54:66-9.  Back to cited text no. 15
Acikgoz, Akduman D, Eskici MZ, Can M, Mungan G, Guven B, et al. Thrombocyte and erythrocyte indices in sepsis and DIC. J Med Biochem 2012;31:61-8.  Back to cited text no. 16


  [Figure 1]

  [Table 1], [Table 2], [Table 3], [Table 4]

This article has been cited by
1 Evaluation of serum and salivary C-reactive protein for diagnosis of late-onset neonatal sepsis: A single center cross-sectional study
Angie Mohamed Samir Tosson,Dina Koptan,Rabab Abdel Aal,Marwa Abd Elhady
Jornal de Pediatria. 2021;
[Pubmed] | [DOI]


Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

  In this article
Materials and Me...
Article Figures
Article Tables

 Article Access Statistics
    PDF Downloaded1449    
    Comments [Add]    
    Cited by others 1    

Recommend this journal