Re-visiting micro ESR as a screening tool for neonatal sepsis

Abstract

Our study evaluated the role of micro-erythrocte sedimentation rate (micro-ESR) in the early detection of neonatal sepsis.

Neonates with >34 completed weeks of gestation, appropriate for gestational age, admitted in our Neonatal Intensive Care Unit with clinical suspicion of early onset sepsis were enrolled in the study. A sepsis screen and blood culture was performed on all the babies within 4 h of admission. The sensitivity of micro-ESR for detecting positive blood culture was calculated and the best cut-off was determined using the Area Under Curve.

Keywords

neonatal disease < disease diagnosis < other bacterial infection < disease

Background

Neonatal sepsis is one of the most common causes of neonatal deaths, especially in low- and middle-income countries, with a case fatality ranging from 25 to 65%.⁽¹⁾ Neonatal sepsis is categorized into early onset (within 72 h of birth) and late (onset after 72 h), with the former contributing over 66% of cases. According to the Delhi Neonatal Infection study (DeNIS), neonatal sepsis was diagnosed in 1945/13,530 (14.3%) neonates, who were admitted to NICU. Out of the total, 840 (6.2%) were culture positive.⁽²⁾ The mortality due to neonatal sepsis can be prevented by timely diagnosis and treatment.

Most neonates with sepsis present with subtle non-specific symptoms and signs. This often leads to difficulty in early and accurate diagnosis^.⁽³⁾ There is no laboratory test that can predict sepsis with 100% sensitivity and specificity. Blood culture, which has been considered the gold standard for confirmation of diagnosis, gives results after 48–72 h and carries the disadvantage of low sensitivity.⁽⁴⁾ Newer biomarkers such as interleukein-6 (IL-6), interleukien-8 (IL-8), highly specific CRP (hs CRP), neutrophil gelatinase-associated lipocalin (NGAL), and cell surface markers CD-11b and sCD163 can detect sepsis within minutes with a sensitivity, specificity approaching 100%.^(5,6) However, the need for special tools and their high cost limit their accessibility in routine clinical practice.

A cost-effective bedside test that can aide in early diagnosis of neonatal sepsis is nonetheless very useful. Total leucocyte count (TLC), absolute neutrophil count (ANC), immature to total lymphocyte ratio (I:T ratio), C-reactive protein (CRP) and micro-Erythrocyte sedimentation rate (micro-ESR) are some haematological parameters utilized, but all have a variable sensitivity and specificity in isolation or in combination.⁽³⁾ Micro-ESR is simple to perform and is a reproducible bed-side test that is frequently used as one of the components of neonatal sepsis screen. It requires only 0.2 mL of blood and gives a result in 1 h but the accuracy and cut-offs have often been questioned. Some researchers have found it to be a good tool,^(7,8) but others have decried it for its low sensitivity and specificity.^(9,10) Thus, the reliability of micro-ESR as a screening test remains uncertain. Most institutes take ‘day of life + 3 mm/1^st h^” to a maximum of 15 mm/1^st h’ as the normal value, corresponding to the 95^th percentile.^(11,12) However, other standards have been recommended.^(13,14)

Methods

Our cross-sectional, diagnostic accuracy study was conducted in the Neonatology unit of department of Paediatrics of a tertiary care teaching hospital over a duration of six months, after obtaining approval from the Ethical Committee of the Institute. Written informed consent was obtained from the parents or the guardians for participation. All neonates of >34 completed weeks of gestation, appropriate for gestational age (AGA), admitted in the Neonatal Intensive Care Unit (NICU) with clinical suspicion of neonatal sepsis within 72 h of birth were enrolled in the study.

Gestational age was calculated in all babies by the date of the last menstrual period (LMP). If this was not available, then a first trimester ultrasound scan was used. In case both were not available, then gestation was defined using physical criteria of modified Ballard's scoring system. Neonates with anaemia (PCV<40) or polycythaemia (PCV>65) and neonates with associated congenital anomaly or clinical suspicion of chromosomal disorders were excluded from the study.

Sepsis was suspected clinically if a neonate presented with hypothermia, lethargy, decreased oral acceptance, apnoea, respiratory distress, fever, or seizures.⁽¹⁴⁾ Sepsis screen, a battery of five tests, named TLC, ANC, I:T ratio, micro-ESR and CRP were performed in all babies within 4 h of admission. Intravenous antibiotics were added in all cases of suspected sepsis according to NICU protocol. Blood cultures were taken before starting antibiotics. Babies with >2 parameters positive were deemed to have a positive sepsis screen. Those with positive blood culture were labeled as confirmed cases of sepsis.

Capillary blood was collected into a standard 75mm heparinized micro-haematocrit tube with internal diameter of 1.1mm, under standard aseptic conditions by heel prick. One end of the tube was sealed with the help of plasticine (1–2 mm) and the tubes were then placed on the stand provided in the micro-ESR kit, after labeling the individual tubes with the patient's particulars, including name and time of blood collection. After 1 h, the distance from the highest point of the plasma column to the highest point of the packed red cell column was measured with the help of a ruler. This figure was increased if the height of the plasma column was found to be more than the day of life + 3 (constant) mm/ after 1^st hour, up to a maximum value of 15mm/1^st hour, irrespective of age. All details were recorded in the case record form.

Considering that micro ESR can detect proven sepsis with sensitivity of 63%, with absolute difference of 10%, and α error of 5%, 89 cases of culture proven sepsis would be required.⁽⁸⁾ Assuming the prevalence of culture proven early onset sepsis in our NICU as 30%, enrolling 267 cases was required. Thus 300 cases of suspected sepsis wre included in our study after accounting for 10% attrition.

Statistical analysis

The collected data was compiled and entered into the SPSS Statistical software Version 20.0. Continuous variables were presented as mean ± standard deviation (SD), and categorical variables were presented as absolute numbers and percentages. Normally distributed continuous variables were compared using the unpaired t-test Categorical variables were analysed using either chi-square test or Fisher's exact test Receiver operating characteristics (ROC) curve were drawn and area under the curve (AUC) was calculated to compare the diagnostic accuracy of micro-ESR at different cut-offs. A p-value of < 0.05 was considered as significant.

Results

Of these 300 cases, 38 (12.7%) were culture positive. Their baseline characteristics were compared to culture negative cases (Table 1).

Table 1.

Baseline characteristics of enrolled cases.

Variable	Culture positive sepsis (n = 38)	Group negative sepsis (n = 262)	P value
Male (n): Female (n)	15:23	115:147
Birth weight (grams)Mean ± SD	2203 ± 731	2170 ± 629	0.772
Gestational age (days)Mean ± S.D.	261 ± 15	262 ± 14	0.653
Day of onset of symptomsMean ± S.D.	1.6 ± 0.6	2.0 ± 0.6	0.630
APGAR at 1 minMean ± S.D.	7.4 ± 1.5	7.4 ± 1.6	0.920
APGAR at 5 minMean ± S.D.	8.3 ± 1.4	8.6 ± 1.3	0.150

Our study confirmed the expected, that mean micro-ESR ; TLC and ANC were significantly higher in the culture positive group as compared to the culture negative group. We calculated sensitivity, specificity, positive and negative predictive values for the various parameters of sepsis screen individually and for the ‘positive sepsis screen’ (defined as >2 positive parameters out of five) (Table 2). ANC had the highest sensitivity (89.4%) whereas the I:T ratio had highest specificity (84.7%). Micro-ESR and TLC had very high negative (93.1%) but low positive predictive values. Further, ROC curve was drawn to determine the best cut-off for micro ESR for determining culture positive sepsis (Figure 1). At cut-off of 1.5 mm after the first hour, micro-ESR showed a sensitivity of 71.4% and a specificity of 62.2% with AUC of 0.598. At cut-off of 2.5 mm, specificity improved to 69.2%, but sensitivity dropped to 35.7%.

Figure 1.

Receiver operating characteristic (ROC) curve showing micro ESR for diagnosis of neonatal sepsis (At cut-off of 1.5mm, sensitivity 71.4%, specificity 62.2%; AUC 0.589).

Table 2.

Comparison of various parameters of sepsis screen in terms of sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV)*.

Variable	Sensitivity (%)	Specificity (%)	PPV (%)	NPV (%)
Micro-ESR	57.9	82.4	32.4	93.1
TLC	68.4	67.6	23.4	93.7
ANC	89.5	35.5	33.0	47.2
I:T RATIO	34.2	84.7	24.5	89.9
CRP	71.0	37.8	14.2	90.0
Sepsis screen	92.1	34.0	16.9	96.7

*Sensitivity = True positive/true positive + false negative; Specificity = True negative/true negative + false positive; PPV = True positive /true positive + false positive; NPV = True negative/true negative + false negative.

Discussion

We observed that micro-ESR had a good negative predictive value, thus could rule out neonatal sepsis, but could not diagnoses early onset sepsis in these neonates. Of the other parameters, none of the cut-offs were able to provide a good sensitivity and specificity for diagnosis of early onset neonatal sepsis. Different cut-offs used in other studies make their interpretation challenging, and almost impossible to compare.

The major strength of our-study was the good sample size of 300 babies. Also, micro ESR was performed by a single observer, who was initially trained and then directly supervised, thus reducing the variability of method. However, the low number of culture positive patients is the major limitation of our study. If we could have used a surrogate marker for diagnosis of sepsis short of blood culture positivity, more clinically useful results would have been obtained.

Micro-ESR has been a part of bed side investigative panel for sepsis screen for decades although without much scientific evidence. Especially when reliable biomarkers are becoming moreavailable, the practice of routinely performing micro-ESR should be re-visited. We recommend that micro-ESR should no longer be employed for the assessment of early neonatal sepsis, as it has no proven clinical utility.

Footnotes

Authors credits

KG, AN: Data collection, writing first draft of manuscript; PB, KG and VB: conceptualized the study, devised its design, and provided critical inputs into manuscript revision. PB and VB: supervised data collection and conduct of study, and provided critical inputs into manuscript revision. All authorapproved the final version of study as submitted, and are willing to be accountable for all aspects of the study.

ORCID iDs

Vikram Bhaskar

Prerna Batra

References

Murthy

Godinho

Guddattu

. Lewis LES, Nair NS. Risk factors of neonatal sepsis in India: a systematic review and meta-analysis. PloS one 2019; 14: e0215683.

Investigators of the Delhi Neonatal Infection Study (DeNIS) collaboration. Characterization and antimicrobial resistance of sepsis pathogens in neonates born in tertiary care centres in Delhi, India: a cohort study. Lancet Glob Health 2016; 4: e752–e760. doi: 10.1016/S2214-109X (16)30148-6. PMID:27633433.

NNF teaching aids: Neonatal sepsis. https://www.newbornwhocc.org/pdf/teaching-aids/neonatalsepsis.pdf. [Accessed 18th June 2021].

Venkatesh

Flores

Luna

, et al. Molecular microbiological methods in the diagnosis of neonatal sepsis. Expert Rev Anti-Infect Ther 2010; 8: 1037–1048.

Taneja

Batra

. Biomarkers as point of care tests (POCT) in neonatal sepsis: a state of science review. J Neonatal Perinatal Med. 2020 Dec 14. doi:10.3233/NPM-200581. Epub ahead of print. PMID: 33337395.

Odabasi

Bulbul

. Neonatal sepsis. Med Bull Sisli Etfal Hospital 2020; 54: 142.

Kutty

Prabhu

Sahana

. Role of micro-ESR in the evaluation of suspected sepsis in neonates. Int J Sci Res 2015; 4: 312–314.

Mondal

Nag

Bandyopadhyay

, et al. Neonatal sepsis: role of a battery of immunohematological tests in early diagnosis. Int J Appl Bas Med Res 2012; 2: 43.

Walliullah

Islam

Siddika

, et al. Role of micro-ESR and I/T ratio in the early diagnosis of neonatal sepsis. MMJ 2009; 18: 56–61.

10.

Desai

Shah

Pandya

, et al. C-Reactive protein, immature to total neutrophil ratio and micro ESR in early diagnosis of neonatal sepsis. IJBAR 2014; 5: 364–366.

11.

Gerdes

. Clinicopathologic approach to the diagnosis of neonatal sepsis. Clin Perinatol 1991; 18: 361–381.

12.

Adler

Denton

. The erythrocyte sedimentation rate in the newborn period. J Pediatr 1975; 86: 942–948.

13.

Diwakar

Rosul

. Revised look at micro erythrocyte sedimentation rate in neonates. Indian Pediatr 1999; 36: 703–705.

14.

Neonatal sepsis. In: AIIMS protocols in Neonatology. 2nd edition. Agrawal R, Deorari A, Paul V, Sankar MJ, Sachdeva A (eds). 2019. Nobel Vision (Medical Book Publishers). Delhi. pp 303-15.