Significance of Platelet Indices as Early Markers for Diagnosis of Neonatal Sepsis
AbdulfatahFaraj Bin Aeshah(1), WafaaFathyElsaaed(2)Asmaa Mohamed HosnyEsh(3),andMohamed Ahmed Arafa(2)
(1) Department of Pediatrics, Faculty of Medicine – Tripoli University, Libya.
(2) Department of Pediatrics, Faculty of Medicine – Zagazig University, Egypt.
(3) Department of Clinical Pathology, Faculty of Medicine – Zagazig University, Egypt.
Corresponding author :AbdulfatahFaraj Bin Aeshah E-mail: [email protected]
ABSTRACT
Background:Neonatal sepsis is major cause of neonatal morbidity and mortality worldwide.
Blood culture and sepsis screening are currently used method, but their utility is limited due to delayed reporting and increased cost. Platelet indices are one such set of parameters which can be helpful in the future diagnosis of neonatal sepsis.This study was aimed to detect the importance of platelet indices as early markers for diagnosis of neonatal sepsis. Patients and methods :This study was a case control study carried out in Neonatal Unit of Pediatric Department at Zagazig university hospitals from 11/2019 to 11/2020, included 132 participants divided into two groups; 1st group included 66 neonates with neonatal sepsis and 2nd group was control group with disease other than sepsis admitted.Peripheral venous blood are collect from all the neonates send for investigations for complete blood count, (PLT, PDW, MPV, P-LCR), CRP, ESR, blood culture. Urine culture. Chest x-ray.Results :There was statistically significant decreased platelets count among the septic cases than control group (281.3±145.9 versus 343.9±134.8) but regarding platelets distribution width (PDW) and P- LCR; they were statistically significant increased among the septic cases than control group (16.6±5.6 versus 11.8±2.1 and 32.3±4.3 versus 21.7±5.7 respectively). P-LCR was the most accurate diagnostic marker (87%) followed by platelets distribution width (PDW) (69.9%) then mean platelet volume (62.5%) and lastly platelets count (48.5%).Conclusion: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.
Keywords: Neonatal sepsis, platelet, Mean platelet volume (MPV), Platelet distribution width (PDW)
INTRODUCTION
Neonatal septicemia is a clinical illness characterized by signs and symptoms of infection in the first month of life, with or without bacteremia. In the newborn critical care unit, sepsis is a prevalent complication. It is caused by a variety of organisms infiltrating the bloodstream, including bacterial, viral, fungal, and protozoalinfections(1).
Neonatal sepsis is a disease condition that depicts the organism's systemic reaction to entering the bloodstream within the first 28 days of life. The incidence of neonatal sepsis is 30 per live birth, according to the National Neonatal Perinatal Database 2002-2003(2).
It is a highly serious illness that is a major cause of newborn morbidity and mortality worldwide; if not recognized and treated promptly, it can result in shock, multiple organ malfunction, permanent impairment, or death(3).
Because clinical signs are imprecise and vague, diagnosing newborn sepsis remains one of the issues that paediatrics faces. As a result, we must rely on investigations to lead us.
Blood culture has historically been considered the gold standard for diagnosing newborn sepsis. Only 20% of symptomatic newborns with suspected earlyonset sepsis (EOS) have a positive blood culture, while only 30% of neonates clinically suspected of having lateonset sepsis (LOS) in a neonatal intensive care unit (NICU) setting have a positive blood culture(4).
Sepsis can damage almost all physiological systems and organs, including the haemostatic system. Clotting cascades do not work in this circumstance, many inflammatory cytokines are generated from endothelium and mononuclear cells, thrombosis comes later, and stimulation of plasminogen and activation of antithrombin occurs in the fibrinolytic system. As a result, fibrinogen and fibrinolytic chemicals are depleted, and clots and bleeding associated with disseminated intravascular coagulation (DIC) ensue(5).
DIC is linked to accelerated platelet breakdown, which results in thrombocytopenia.
Thrombocytopenia is commonly observed prior to the bacteria being cultivated from the blood. Furthermore, bacteria or their products may cause endothelium injury, resulting in platelet adhesion and aggregation, or may bind directly to platelets, resulting in aggregation and hastened removal from blood circulation(1).
The usual ranges of platelet counts in neonates are 150.000- 400.000/L, which are similar to adult values. Platelet counts are fewer than 80.000/L in approximately 40% of individuals with severe sepsis. Mean platelet volume (MPV) refers to the average size of platelets and is used as a metric of platelet production rate and activation. MPV values are typically between 8.5fl to 12.5fl(6).
In destructive thrombocytopenia, MPV levels rise while in hypoproliferative thrombocytopenia, they fall. Platelet distribution width (PDW) is a measure of platelet size variance. PDW levels in the normal range are between 10% and 17%. When turnover is enhanced, PDW levels increase in platelet consumption, and the behavior is comparable to MPV during acute severe infections(4).
Platelet-large cell ratio (P-LCR) denotes the proportion of platelets larger than 12 fL in the overall platelet count, and the typical range is less than 30%. These indices can be measured using a low-cost, widely available routine blood count(7). The use and use of MPV, PDW, and P-LCR in newborn sepsis is uncertain, and data on how sepsis affects these indices is inconsistent.So the aim of this study was to demonstrate the value of platelet indices in monitoring early diagnosis of neonatal sepsis. Early diagnosis is essential to reduce morbidity and mortality.
Patients AND METHODS
This case control study carried out from 11/2019 to 11/2020 in Neonatal Unit of Pediatric department at Zagazig university hospitals. included 132 participants divided into two groups;
each group included (66) patients.1st group with neonatal sepsis with (44 males, 22 females)
ages (1-20) years old and the mean age (5.7±5.1) was years and 2nd group was control group without sepsis admitted with (34 males, 32 females) ages (1-10) years old and the mean age (2.4±1.6).
Inclusion criteria: Gestational age ≥ 32 weeks. Post natal age from birth to 28 days.
Neonatesaverageweight≥1.5kg.Anysuspectedcaseofneonatalsepsiswithmaternalrisk
factor for sepsis e.g prolonged labour, PROM, maternal intra partum fever, UTI and chorioaminionitis. Neonates with clinical sepsis (apnea, poor suking, reduced activity, lethargy, hypotonia, hyporeflexia, cyanosis, respiratory distress, irritability, hypothermia, hyperthermia, not doing well neonates). Neonates who suffer from diseases other than suspectedsepsisasjaundice,TTN,meconiumaspirationetc…,historyofchorioamnionitis Exclusion criteria: Newborn above 28 days. Absence of informed consent. Neonates who received antibiotics before admission.
All patients full history taking, clinical examination (physical examination, vital signs, anthropometric measurement). Laboratory Investigations included;
Complete blood count (CBC); 1 ml blood samples were taken from all the patients and controls and collected in tubes containing EDTA. The complete blood counts (CBCs) performed using a Sysmex XT1800i analyser (Germany) (2).
C-Reactive Protein (CRP); samples were directly collected into EDTA-coated tubes. Plasma wasobtainedbycentrifugationat4°Cfor15minat1000xg,andstoredat−80C°until use.
Freezing/thawing cycles were carefully avoidedse(7). Blood culture:
Blood samples were taken at time of presentation of sepsis. Aerobic and anaerobic cultures were done on blood agar plates at 10% Co2 and on MacConkey agar plates. Isolated colonies were further identified by examination of their colony morphology, gram stained smears, biochemical and enzymatic reactions. True bacteremia was considered when the blood culture was positive within 72 hours. If no growth was detected, the sample was incubated up to 10 days with further subcultures every other day on solid media. If no growth appeared after 10 days of incubation, blood culture was considered negative. Antibiotic sensitivity test was done by Kirby BaurTechnique(8).
Urine culture: Urine sample collected by supra pubic aspiration. All the samples were sent to the laboratory within a half-hour of the procedure(9).
L.P : if needed;Cerebrospinal fluid examination should be done in all neonates with suspected meningitis, CSF sent cultures to diagnose Bacteria, fungi, and viruses meningitis(10). Statistical Analysis
Collected data were recorded then presented and analyzed statistically by computer using SPSS version 22 (SPSS Inc. Chicago, IL, U.S.A). P value equal to or less than 0.05 was considered statistically significant.
RESULTS
Table 1, showed that streptococcus agalactiae causing pneumonia was the commonest organism (30.3%) among septic cases followed by klebsilla and staph epidermidis; each of
them (24.2%) then pseudomonas aeruginosa (9.1%) and lastly staph hominis and candida (6.1%) for each.
Table (1): Frequency distribution and percentage of isolated organisms among the septic cases:-
Isolated organisms
The septic cases No(66) % Klebsilla
Pseudomonas aeruginosa
16 6
24.2%
9.1%
Staph epidermidis Staph hominis
Streptococcus agalactiae (pneumonia)
16 4 20
24.2%
6.1%
30.3%
Fungal infection(Candida) 4 6.1%
Figure 1,2, this study showed there was statistically significant decreased platelets count among the septic cases than control group (281.3±145.9 versus 343.9±134.8) but regarding platelets distribution width (PDW) and P-LCR; they were statistically significant increased among the septic cases than control group (16.6±5.6versus 11.8±2.1 and32.3±4.3 versus21.7±5.7 respectively). In regard to mean platelet volume (MPV), there was no statistically significant difference between the septic cases and control group.
Fig (1); Bar chart for platelets distribution width (PDW) and P-LCR between the case and control groups
0 5 10 15 20 25 30 35
Platelets distribution width P-LCR 16.6
32.3
11.8
21.7
case control
Fig (2); Bar chart for platelets count between the case and control groups
Figure 3; showed that A) P-LCR were the most accurate diagnostic marker (87%), followed by B)platelets distribution width (PDW) (69.9%), then mean C) platelet volume (62.5%) and lastly D)platelets count (48.5%).
Fig (A); ROC curve for the diagnostic ability of P- LCR in the prediction of neonatal sepsis
Fig (B); ROC curve for the diagnostic ability of PDW in the prediction of neonatal sepsis:
0 100 200 300 400 500 600
Case Control
Platelets count
Fig (C); ROC curve for the diagnostic ability of MPV in the prediction of neonatal sepsis:
Fig (D); ROC curve for the diagnostic ability of platelets count in the prediction of neonatal sepsis
Table 2; this study showed that (100.0%) of the septic cases had high CRP while only (18.2%) of control group had high CRP. there was statistically significant difference between the septic cases and control group regarding P-LCR where (97.0%) of the septic cases had elevated P-LCR while only (33.3%) of control group had elevated P-LCR. There was statistically significant difference between the septic cases and control group regarding mean platelet volume where (78.8%) of the septic cases had elevated MPV while (48.5%) of control group had elevated MPV. There was statistically significant difference between the septic cases and control group regarding platelets distribution width (PDW) where (87.9%) of the septic cases had elevated PDW while (45.5%) of control group had elevated PDW. There was no statistically significant difference between the septic cases and control group regarding platelets count where (18.2% and 15.2%) of the septic cases and control group respectively had decreased platelets count.
Table (2): Relation between sepsis and (CRP, P-LCR, MPV, PDW platelets count) according to the cut off value between septic group and control group:-
Septic group Control group
χ² p-value Odds
(95% CI)
No(66) % No(66) %
CRP
>6 (78) 66 100.0% 12 18.2%
FET 0.001** 0.15
(0.09%-0.26%)
< 6 (54) 0.0 0.0% 54 81.8%
P-LCR
>25.6% (86) 64 97.0% 22 33.3% FET 0.001 64
< 25.6% (46) 2 3.0% 44 66.7% ** (14.3%-28.6%) MPV
>10.1(fl) (86) 52 78.8% 32 48.5%
13.9 0.001** 3.9
(1.8%-8.5%)
< 10.1(fl) (46) 14 21.2% 34 51.5%
PDW
>11.8% (88) 58 87.9% 30 45.5%
26.7 0.001** 8.7
(3.6%-21.1%)
< 11.8% (44) 8 12.1% 36 54.5%
Platelets count
<150(/μ L) (88) 12 18.2% 10 15.2%
0.2 0.6 1.2
(0.49%-3.1%)
> 150(/μ L) (44) 54 81.8% 56 84.8%
**FET=Fisher Exact test ** MPV: Mean platelet volume
**Statisticallyhighlysignificantdifference(P≤0.001)** χ²: chi square test **** χ²: chi square test
Table 3; this study showed that there was statistically significant negative correlation between platelets count with mean platelet volume and P-LCR (increased platelets count is associated with decreased in mean platelet volume and P-LCR) among the septic cases.
Regarding other variables, there was no statistically significant correlation with platelets count among the septic cases.
Table (3): Correlation between platelets count with patients' characteristics and laboratory investigations among the septic group:
Variable
Platelets count
r^ p SIG Age 0.01 >0.05 NS Gestational age 0.03 >0.05 NS WBCs (/μ L) 0.01 >0.05 NS CRP 0.02 >0.05 NS PDW (%) -0.1 >0.05 NS
MPV (fl) -0.3 0.01* S P-LCR (%) -0.3 0.004* S
** Statistically significant difference (P ≤ 0.05)
**SIG: significant
**r: correlation
DISCUSSION
Streptococcus agalactiae, which causes pneumonia, was the most common isolate (30.3%), followed by klebsilla and staph epidermidis (24.2 %), then pseudomonas aeruginosa (9.1%), and finally Staph hominis and Candida (6.1 %).
In terms of mean platelet volume (MPV), there was no statistically significant difference between the septic cases and the control group, which was consistent with the findings of Tayman et al. (6), who found that platelet count, MPV, and PDW were significantly different between sepsis and control groups (270.898 versus 24059, 8.30.3 versus 10.30.9, and 16.60.2 versus 17.30.1, p-value0.05).
Also, Madani et al. (2) agreed with this finding, concluding that in the sepsis group, MPV, PDW, and P-LCR were significantly higher than in the control group (10.30.23 versus 9.30.19, 13.50.5 versus 11.70.27, and 28.281.44 versus 21.591.16, p-value=0.005, 0.005, and 0.001), but this was in contrast with our finding regarding platelet count
Finally, Mittal et al.(4)concluded that the case group had a bigger number of participants with higher platelet indices than the control group (P 0.05), and the mean platelet count was lower in the case group (1.098 0.747 lakhs/mm3) than the control group (2.038 0.762 lakhs/mm3; P 0.0001). The mean MPV in the case group was greater (11.85 1.716 fl) than in the control group (9.81 1.460 fl), P 0.0001. Furthermore, the mean PDW value in the case group was greater (20.68 2.239) than in the control group (18.69 1.974), P 0.0001.
In terms of platelet indices' diagnostic abilities, the current study discovered that P-LCR was the most accurate diagnostic marker (87 % ), followed by platelets distribution width (69.9 %), mean platelet volume (62.5 %), and platelets count (62.5 %). (48.5 %). This was similar to the findings of Madani et al.(2), who discovered that P-LCR was the most accurate diagnostic marker, followed by platelets distribution width (PDW) and mean platelet volume, with sensitivity and specificity of (65.2%& 80.0%, 70.0%&70.5%, and 65.3%&75.0%, respectively).
In contrast, Mittal. et al. (4) assessed the sensitivity and specificity of platelet indices for the diagnosis of neonatal sepsis by comparing them to blood culture, which was the gold standard for the diagnosis of neonatal sepsis, and discovered that thrombocytopenia was the most sensitive marker (83.08%), followed by MPV and PDW in detecting babies with sepsis.
It did, however, have a low specificity (20.33%). However, when MPV and PDW were combined, or when all three markers (MPV + PDW + PC) were merged, the specificity climbed to 46.34 %. On ROC curves, the area under the curve (AUC) for platelet count was 0.156, 0.814 for MPV, and 0.743 for PDW.
Concerning elevated platelet indices, the following study discovered a statistically significant difference between the septic and control groups in terms of elevated P-LCR, mean platelet volume, and platelets distribution width (PDW), with 97.0 %, 78.8 %, and 87.9
% of the septic cases having elevated P-LCR and only (33.3 %, 48.5 %, and 45.5 %) of the control cases having elevated P-LCR. This was consistent with the findings of Karne et al. (12), who discovered that out of all sepsis positive patients, severe thrombocytopenia was present in 57.5 %, while mild and moderate thrombocytopenia were present in 22.5 % and 20.0%, respectively. In non-infected neonates,74.5 % had moderate thrombocytopenia and 14.9 %
had severe thrombocytopenia at the same time. Arif SH et al. (11) found thrombocytopenia in 83.5 % of neonates with sepsis.
In terms of PDW, Karne et al. (12) discovered that 45 out of 103 sepsis cases (43.69 %) had an increased PDW value, and Patrick &Lazarchick.(13) found that there was a significantly increased presence of bacteremia in neonates with MPV greater than 10.8f L and/or PDW greater than 19.1%.
The following study discovered a statistically significant negative link between platelets count and mean platelet volume and P-LCR among septic cases (increasing platelets count is associated with lower mean platelet volume and P-LCR). In terms of other variables, there was no statistically significant link between platelet count and septic cases. This was congruent with the findings of Tayman et al (6), who discovered that platelet count was connected negatively with MPV (Rho = - 0.852, p = 0.01) and PDW (Rho = - 0.765, p = 0.014), and positively with MPV and PDW (Rho = 0.791, p = 0.022).
Conclusion:
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, as part of a routine automated cell count, are cheap and readily available tests that can be used as valuable clues in the diagnosis of neonatal sepsis.
REFERENCES
1- Choudhary R.R, Makwana M, Mourya H.K,.et al., (2018); Evaluation of platelet and its indices as a marker of neonatal sepsis: a prospective case control study. IJCP;
5(5):1898-903.
2- Madani S. H., Amiri S., Khazaei S., et al., (2019): platelet indices as useful indicators of neonatal sepsis. Journal of Evolution of Medical and Dental Sciences, 8(20), 1612- 1618.
3- Grant C.H, Arnott A., Brook T.,et al., (2018): Reducing antibiotic exposure in suspected neonatal sepsis. Clinical Pediatrics (Phila); 57(1):76-81.
4- Mittal A., Arya S., Charan L. S., et al., (2020): Evaluation of platelet indices as additional diagnostic tool for neonatal sepsis. Astrocyte, 4(4), 205
5- González-Vázquez P., Larrañeta E., McCrudden M.T.C,.et al., (2017): Transdermal delivery of gentamicin using dissolving microneedle arrays for potential treatment of neonatal sepsis. Journal of Controlled Release; 265: 30-40.
6- Tayman, C., Tonbul, A., Akca, H., et al., (2014): mean platelet volume (MPV) may simply predict the severity of sepsis in preterm infants. Clin. Lab, 1
7- Chiarot E, Spagnuolo A and Maccari S., et al., (2018); Protective effect of Group B Streptococcus type-III polysaccharide conjugates against maternal colonization, ascending infection and neonatal transmission in rodent models. Scientific Reports;
8(1):2593.
7- Giuliani A. L., Berchan M., Sanz, J. M.,et al.,(2019)The P2X7 receptor is shed into circulation: correlation with C-reactive protein levels. Frontiers in Immunology, 10, 793
8- Tadesse M. M., Matebie Z. A., Tullu K. D., et al., (2019):Multi-drug resistant, extended spectrum beta-lactamase and carbapenemase-producing bacterial isolates among children under five years old suspected bloodstream infection in a specialized Hospital in Ethiopia: Cross-sectional study
9- Manandhar S. R., and Basnet R. (2020):Micro-Erythrocyte Sedimentation Rate in Neonatal Sepsis of a Tertiary Hospital: A Descriptive Cross-sectional Study. Journal of the Nepal Medical Association, 58(226).
10- Pearson J, and Fuller G. (2012):"Lumbar punctures and cerebrospinal fluid analysis". Medicine. 40 (8): 459–462.
11- Arif S.H, Ahmed. I, Ali SM et al (2012):Thrombocytopenia and bacterial sepsis in neonates. Indian J Hematol Blood Transfusion.; 28 (3):147–51.
12- Karne T. K., Joshi D.D., Zile U., et al., (2017); Study of platelet count and platelet indices in neonatal sepsis in tertiary care institute. MVP Journal of Medical Science; 4(1), 55-60.
13- Patrick C.H, and Lazarchick J. (2016):The effect of bacteremia on automated platelet measurments in neonates. Am J ClinPathol.; 93(3):391–4.
