Effectiveness of zinc protoporphyrin/heme ratio and ferritin for assessing iron status in preterm infants

Abstract

BACKGROUND:

Since iron is crucial for many tissue processes, we, therefore, aimed to assess ferritin and the zinc protoporphyrin to heme ratio (ZnPP/H) as indicators of iron status in preterm newborns, particularly during certain inflammatory episodes.

METHODS:

From 170 preterm babies, paired ferritin and ZnPP/H measurements were collected twice (on the first postnatal day and six weeks later). To compare these measures and assess the impact of anemia, sepsis, and packed red blood cell transfusion (PRBT), three different scenarios were considered.

RESULTS:

Compared to the non-anemic group, the anemic patients’ serum ferritin level was considerably lower (p = 0.044), whereas the anemic patients’ ZPP/H ratio was significantly greater (p < 0.001). In neonates with sepsis, ferritin levels were considerably greater in both anemic and non-anemic septic neonates compared to neonates without sepsis (p < 0.001 for each). Regarding ZPP/H ratio, no appreciable variations were found between the two groups. In addition, serum ferritin significantly increased following each PRBT (p < 0.001 for each). As a result of each PRBT, the ZPP/H ratio considerably decreased (p < 0.001).

CONCLUSION:

As a measure of iron status during particular inflammatory processes like infection and PRBT, ZnPP/H may be more accurate.

Keywords

Ferritin PRBT preterm ZPP/H

1 Introduction

Due to irreparable neurodevelopmental delays caused by iron insufficiency during important windows of brain development, iron sufficiency in the neonatal population is of great clinical importance [1 –3]. Extremely low gestational age neonates (ELGANs) are at risk of iron insufficiency since the majority of iron transfer happens late in pregnancy. Iron consumption is additionally increased by rapid expansion and phlebotomy [4]. Although iron has strong antioxidant effects and can be harmful when present in excess, neonatal caregivers may be hesitant to start iron supplementation in preterm newborns [5]. Both European and American recommendations state that elemental iron supplements of 2-3 mg/kg/day should be given to all premature newborns [6, 7]. Preterm newborns are at significant risk for both iron insufficiency and iron toxicity due to individual demands and enteral absorption being diverse [7], It seems necessary to measure iron status to personalize iron supplements.

Serum ferritin (SF) and the zinc protoporphyrin-to-heme ratio (ZnPP/H) are two frequent indicators of iron sufficiency in newborns. Plasma ferritin has been utilized as an indirect indicator of iron sufficiency because ferritin binds to and stores iron intracellularly and serves as an iron carrier in plasma [8]. However, the confounding effects of illness, inflammation, and malnutrition limit its interpretation in developing nations. The primary drawback of SF is its acute phase response, which can conceal an infection’s iron insufficiency [9]. Iron incorporation is insufficiently measured by the erythrocyte zinc protoporphyrin/heme (ZnPP/H) ratio. Iron and protoporphyrin IX combine to create heme [10]. When there is an insufficient supply of iron, zinc takes its position in the protoporphyrin ring [10]. Concerning the erythrocyte’s moles of heme, the intrinsic zinc protoporphyrin fluorescence is quantified and expressed as a ratio [11]. ZnPP/H ratios reveal iron deficit earlier than traditional assessments of tissue iron reserves because they measure incomplete iron incorporation [10]. ZnPP/H is more sensitive in young infants than hemoglobin or plasma ferritin level (a marker of storage) (a measure of anemia) [12].

The purpose of this study was to elucidate the significance of ferritin and the zinc protoporphyrin to heme ratio (ZnPP/H) as indicators of iron status in preterm newborns and to investigate how particular inflammatory events can influence these measurements.

2 Subjects and methods

2.1 Study design and population

This cohort study was conducted on 170 preterm VLBW neonates aged ≤34 weeks gestation admitted at the Neonatal Intensive Care Units (NICUs) of the Pediatric Department at Benha University Hospital between January 2021 and August 2022. The study protocol was approved by the Ethical Scientific Committee of the Faculty of Medicine, Benha University conferring to the World Medical Association Declaration of Helsinki [13], and informed consent was obtained from parents/guardians before enrollment in the study. Inborn errors of metabolism, hematological diseases, major congenital or chromosomal disorders, and full-term newborns were excluded. Neonates were specifically excluded if the maternal Hct <30; or if the mothers had been identified as having preeclampsia, chorioamnionitis, or insulin-dependent diabetes mellitus (IDDM). Every newborn was given a history to go along with a clinical examination that focused on any pertinent clinical signs of sepsis and necrotizing enterocolitis (NEC). Newborn erythrocyte indices, hematocrit, and blood hemoglobin concentration reference ranges were used to make the diagnosis of neonatal anemia [14, 15]. When the neonates reached full enteral feeding (100 mL/kg), iron supplementation (2 mg/kg) was started. We withheld iron supplementation if serum ferritin levels were >350 ng/mL. According to unit policy, the number and volume of packed red blood cell (PRBC) transfusions were documented. The cutoff for giving PRBC was a hematocrit of 36%, oxygenation of >35%, or a mean airway pressure of >6 to 8 cmH2O by positive pressure ventilation, as well as a hematocrit of 31% with respiratory support, oxygen therapy, or recurrent apneic episodes occurring more than 9 times per 12 hours. The amount of erythrocyte transfusions varied between 10 and 15 mL/kg depending on the infants’ clinical status. The indicator for a laboratory assessment of iron status within 7 days of a recent red blood cell transfusion has been established [16]. Sepsis was defined as a culture-positive infection either two days before or one day after laboratory markers of iron status were obtained [14]. Using modified Bell’s staging criteria [17], infants with necrotizing enterocolitis (NEC) were detected two days before or one day after laboratory markers of iron status were obtained. ZnPP/H and ferritin were assessed twice, on the infant’s first day of birth and six weeks later, along with the infant’s age and corrected gestational age at the time of laboratory testing. Three newborns with NEC died and were eliminated from the study.

2.2 Laboratory investigations

Venous blood was collected from studied neonates on 1st day of life and after 6 weeks to assess complete blood count (CBC) with differential leukocyte count to detect Immature to Total (I/T) neutrophil ratio. Neonatal sepsis was evaluated on 1st day of life and after 6 weeks by blood, culture using BD BactecTM and BD PhoenixTM 100 systems and serum C- reactive protein (CRP) level.

Iron status was assessed by measuring serum ferritin and zinc protoporphyrin (ZPP) levels on 1st day of life, after 6 weeks, and after PRBC in anemic neonates. Commercial sandwich enzyme-linked immunosorbent assays (ELISA) were used for human ferritin (Cat. # ab108698, Abcam, UK) with a kit sensitivity of 0.53 ng/ml and recovery 98.66 % and for human ZPP (Cat. # MBS167672, MyBioSource, USA). The kit sensitivity is 0.24 ng/ml and linearity is 0.5–200 ng/ml. Ferritin concentration is age and sex-dependent thus its expected values are variable. ZPP values were measured as μg/dL (ng/ml ÷ 10) and converted to μmol/mol heme (ZPP/H ratio) by calculating with the following equation [Stanton NV, Gunter EW, Parsons PJ, Field PH. Empirically determined lead-poisoning screening cutoff for the Protofluor-Z hematofluorometer. Clin Chem 1989; 35:2104-7]. $\begin{matrix} ZPP / H ratio & = & \frac{μ mol of ZPP}{mol of heme} \\ = & \frac{μ g of ZPP}{g of Hgb} \times 25.8 \end{matrix}$

2.3 Statistical analysis

We used IBM SPSS Statistics 24.0 (IBM Corp., Armonk, NY, USA) for statistical analyses. Patient characteristics were described as mean, SD for normally distributed data, or median and interquartile range (IQR) for skewed data. Shapiro-Wilk and Levine’s tests were used to detect normality in each group and homogeneity of variances respectively. The frequency with percentage (%) was used for presenting qualitative data. Pearson Chi-square (χ²) test was used to analyze the differences between groups. Student-t and Mann-Whitney U tests were used to test for significant differences between two groups for parametric and non-parametric data, respectively. While one-way ANOVA and Kruskal-Wallis tests were used when more than two groups were tested. Paired t-test and Wilcoxon signed-rank test were used to compare the means of two samples on 1st day of life and after 6 weeks, in data with normal distribution and abnormal distribution, respectively. Receiver operating characteristic (ROC) analysis was conducted to assess the performance (accuracy) and effectiveness (sensitivity, specificity) of studied parameters to discriminate between diseased and non-diseased neonates at the optimal cutoff value. In all applied tests, P-values<0.05 were considered significant.

3 Results

In the current study, there were 170 neonates, 88 of whom were male (51.8%) and 82 of whom were female (48.2%). The mean gestational age was 32.5±1.9 weeks, 16.5% of the cases were <1500 gm in weight and 4.7% were small for gestational age.

When it comes to the history of maternal risk factors, 120 neonates (70.6%) had a history of prenatal steroids, and premature membrane rupture was reported in 75 neonates (44.1%). 39 newborns (22.9%) were born vaginally, compared to 141 neonates (77.1%) who were delivered via cesarean section.

Out of the cases that were studied, 81 neonates (49.1%) had anemia at the age of 6 weeks and 78 neonates (47.3%) had sepsis that was determined to be present by a positive blood culture. Of these, 47 neonates (28.4%) required packed red blood cell transfusions once, 14 neonates (8.5%) required packed red blood cell transfusions twice, and 9 neonates (5.4%) required packed red blood cell transfusions three times.

Hemoglobin, HCT, MCHC, MCV, and reticulocytes were found to have considerably decreased after 6 weeks (p < 0.001 for each), while IT ratio, CRP, serum ferritin, and ZPP/H ratio were found to have significantly increased after 6 weeks (p < 0.001, p = 0.032, p = 0.002, and p < 0.001 respectively). However, there was no discernible difference between the TLC count and platelet count Table 1.

Table 1
Comparing CBC parameters, ferritin, and ZPP/H of the studied group on 1st post-natal day and 6 weeks later

1st post-natal day After 6 weeks Test P

N = 170 N = 165^*

Hb (g/dl) Mean±SD 15.8±2 13.2±2.5 Pt = 12.1 <0.001

Range 11–20.5 5.8–18.3

HCT (%) Mean±SD 47.5±5.6 40.3±8.13 Pt = 11.1 <0.001

Range 36.9–62 21–59.9

MCHC (g/dl) Mean±SD 25.7±6.7 23.8±6.2 Pt = 7.8 <0.001

Range 22.7–35.7 20–37.6

MCV (fL) Mean±SD 89.5±21.5 82.3±15.4 Pt = 6.7 <0.001

Range 79.7–120.4 75–122

Platelets (×10³/L) Mean±SD 228±98 249±126 Pt = 1.8 0.077

Range 40–495 40–708

TLC (×10³/L) Mean±SD 12.02±4.9 11.8±5.8 Pt = 0.61 0.54

Range 2.6–29.6 3.7–46.2

Reticulocytes (%) Mean±SD 6.9±2.3 2.1±1.9 Pt = 6.3 <0.001

Range 3.2–10.3 1.1–6.2

IT Ratio Mean±SD 0.05±0.05 0.12±0.09 Pt = 8.6 <0.001

Range 0–0.26 0–0.43

CRP (mg/L) Median 6 12 W = 2.1 0.032

Range 1–48 1–139

Serum ferritin (ng/ml) Median 51.4 87.3 W = 4.2 <0.001

Range 16.8–183.7 26.2–246.3

Iron (μg/dL) Mean±SD 72.4±54.1 74.3±37.5 Pt = 0.61 0.543

Range 45.8–164.7 46.9–182.3

TIBC (μg/dL) Mean±SD 161.5±42.6 195.1±68.5 Pt = 0.61 0.621

Range 91–193.8 94.8–235.9

ZPP/H (μmol/mol) Mean±SD 84.6±18.4 114.2±26.8 Pt = 9.5 <0.001

Range 53.4–121.6 62.3–166.5

		1st post-natal day	After 6 weeks	Test	P
Hb (g/dl)	Mean±SD	15.8±2	13.2±2.5	Pt = 12.1	<0.001
	Range	11–20.5	5.8–18.3
HCT (%)	Mean±SD	47.5±5.6	40.3±8.13	Pt = 11.1	<0.001
	Range	36.9–62	21–59.9
MCHC (g/dl)	Mean±SD	25.7±6.7	23.8±6.2	Pt = 7.8	<0.001
	Range	22.7–35.7	20–37.6
MCV (fL)	Mean±SD	89.5±21.5	82.3±15.4	Pt = 6.7	<0.001
	Range	79.7–120.4	75–122
Platelets (×10³/L)	Mean±SD	228±98	249±126	Pt = 1.8	0.077
	Range	40–495	40–708
TLC (×10³/L)	Mean±SD	12.02±4.9	11.8±5.8	Pt = 0.61	0.54
	Range	2.6–29.6	3.7–46.2
Reticulocytes (%)	Mean±SD	6.9±2.3	2.1±1.9	Pt = 6.3	<0.001
	Range	3.2–10.3	1.1–6.2
IT Ratio	Mean±SD	0.05±0.05	0.12±0.09	Pt = 8.6	<0.001
	Range	0–0.26	0–0.43
CRP (mg/L)	Median	6	12	W = 2.1	0.032
	Range	1–48	1–139
Serum ferritin (ng/ml)	Median	51.4	87.3	W = 4.2	<0.001
	Range	16.8–183.7	26.2–246.3
Iron (μg/dL)	Mean±SD	72.4±54.1	74.3±37.5	Pt = 0.61	0.543
	Range	45.8–164.7	46.9–182.3
TIBC (μg/dL)	Mean±SD	161.5±42.6	195.1±68.5	Pt = 0.61	0.621
	Range	91–193.8	94.8–235.9
ZPP/H (μmol/mol)	Mean±SD	84.6±18.4	114.2±26.8	Pt = 9.5	<0.001
	Range	53.4–121.6	62.3–166.5

Data represented as mean±SD or median and range. ^*Five cases were expired before completing the study (excluded), Pt: Paired t-test, W:Wilcoxon signed-rank test, CBC; complete blood count, HCT; hematocrit, MCV; Mean corpuscular volume, MCHC; Mean Corpuscular Hemoglobin Concentration, TLC; total leucocytic count, IT ratio; immature to total neutrophils ratio, CRP; C- reactive protein, TIBC; total iron binding capacity, ZPP/H ratio; zinc protoporphyrin to hemoglobin ratio. Bold values are significant at p≤0.05.

At 6 weeks postnatal age, the studied neonates were divided into an anemic group (81 neonates) and a non-anemic group (84 neonates), and 5 neonates were expired. A comparison of laboratory parameters revealed that hemoglobin, HCT, MCHC, and MCV were significantly lower in the anemic than non-anemic group (p < 0.001 each), and also serum ferritin was significantly lower in the anemic than non-anemic group (p = 0.044), while ZPP/H ratio was significantly higher in anemic than non-anemic patients (p < 0.001), while no significant differences between both groups were detected regarding other CBC parameters or CRP Table 2.

Table 2

Comparing CBC parameters, ferritin, and ZPP/H of the studied group regarding anemia at 6 weeks postnatal age

		Anemic group	Non-anemic group	Test	P
		N = 81	N = 84
Hb (g/dl)	Mean±SD	11.3±1.4	14.8±1.7	t = 12.1	<0.001
	Range	7–11.5	12–20
HCT (%)	Mean±SD	32.7±5.4	44.3±6.4	t = 8.7	<0.001
	Range	21.9–40.4	40–59.5
MCHC (g/dl)	Mean±SD	22.8±5.1	33.8±6.7	t = 7.4	<0.001
	Range	21.9–30.1	30–37.6
MCV (fL)	Mean±SD	80.7±8.4	112.3±5.4	t = 7.6	<0.001
	Range	70.2–93.5	95–122
Platelets (×10³/L)	Mean±SD	265±150	238±135	U = 1.2	0.24
	Range	40–700	41–650
TLC (×10³/L)	Mean±SD	12.5±6.6	11.5±4.6	U = 2.5	0.06
	Range	3.7–36.1	3.7–35.5
Reticulocytes (%)	Mean±SD	4.8±1.2	3.9±1.3	U = 0.68	0.49
	Range	2.8–9.2	2.1–7.4
IT Ratio	Mean±SD	0.12±0.1	0.09±0.09	U = 1.2	0.22
	Range	0–0.43	0–0.42
CRP (mg/L)	Median	8	6	U = 2.2	0.069
	Range	1–139	1–96
Serum ferritin (ng/ml)	Median	44.5	73.6	U = 2.9	0.044
	Range	22.3–165.0	16.8–248.3
Iron (μg/dL)	Mean±SD	70.69±41.33	66.2±34.5	t = 0.86	0.640
	Range	42–139.9	38.3–156.9
TIBC (μg/dL)	Mean±SD	169.7±52.1	182.8±74.1	t = 0.96	0.686
	Range	79.9–189.5	86.6–199.3
ZPP/H (μmol/mol)	Mean±SD	115.2±33.4	85.6±21.3	t = 11.1	<0.001
	Range	71.2–207.5	43.8–153.2

Data represented as mean±SD or median and range. t: Student t-test; U; Mann-Whitney U test. CBC; complete blood count, Hb; hemoglobin, HCT; hematocrit, MCV; Mean corpuscular volume, MCHC; Mean Corpuscular Hemoglobin Concentration, TLC; total leucocytic count, IT ratio; immature to total neutrophils ratio, CRP; C- reactive protein, TIBC; total iron binding capacity, ZPP/H ratio; zinc protoporphyrin to hemoglobin ratio. Bold values are significant at p≤0.05.

Our patients were divided into two groups based on blood culture results at six weeks postnatal age: the septic group (78 newborns) and the non-septic group (87 neonates).

Neonates with sepsis had considerably lower platelet counts than patients without sepsis (p < 0.001). While the sepsis group’s serum ferritin, CRP, and IT ratio were all noticeably higher (p < 0.001 for each). Regarding the other CBC parameters and ZPP/H ratio, no appreciable variations between the two groups were detected Table 3.

Table 3

Comparing CBC parameters, ferritin, and ZPP/H of the studied group regarding sepsis at 6 weeks postnatal age

		Septic group	Non-septic group	Test	P
		N = 78	N = 87
Hb (g/dl)	Mean±SD	12.8±2.4	13.5±2.6	t = 1.8	0.059
	Range	7–18.5	8–20
HCT (%)	Mean±SD	39±7.4	40.8±6.4	t = 1.4	0.15
	Range	24–57	21.9–59.9
MCHC (g/dl)	Mean±SD	25.8±5.1	26.8±6.7	t = 0.41	0.65
	Range	21.9–36.1	22–37.6
MCV (fL)	Mean±SD	90.7±8.4	92.3±5.4	t = 0.63	0.54
	Range	70.2–112	75–122
Platelets (×10³/L)	Mean±SD	192±104	297±124	U = 5.9	<0.001
	Range	40–440	95–708
TLC (×10³/L)	Mean±SD	12.6±4.4	11.5±6.6	U = 1.1	0.25
	Range	3.7–36.1	4.6–15.5
Reticulocytes (%)	Mean±SD	3.9±1.7	3.6±2.3	U = 1.4	0.15
	Range	2.9–5.4	2.4–6.1
IT Ratio	Mean±SD	0.14±0.1	0.08±0.07	U = 3.1	<0.001
	Range	0.11–0.43	0–0.12
CRP (mg/L)	Median	12	6	U = 6.4	<0.001
	Range	1–139	1–48
Serum ferritin (ng/ml)	Median	124	75	U = 8.6	<0.001
	Range	16.8–662.3	16.2–316.1
Iron (μg/dL)	Mean±SD	76.9±57.8	72.9±51.4	t = 0.43	0.278
	Range	48.9–145.9	47.7–165.7
TIBC (μg/dL)	Mean±SD	188.64±67.2	168.6±43.1	t = 0.62	0.501
	Range	89.9–305.8	92.4–219.8
ZPP/H (μmol/mol)	Mean±SD	114.3±26.5	123.8±28.5	U = 2.1	0.062
	Range	74.5–209.1	82.7–227.8

Data represented as mean±SD or median and range. t: Student t-test, U; Mann-Whitney U test. CBC; complete blood count, Hb; hemoglobin, HCT; hematocrit, MCV; Mean corpuscular volume, MCHC; Mean Corpuscular Hemoglobin Concentration, TLC; total leucocytic count, IT ratio; immature to total neutrophils ratio, CRP; C- reactive protein, TIBC; total iron binding capacity, ZPP/H ratio; zinc protoporphyrin to hemoglobin ratio. Bold values are significant at p≤0.05.

The anemic group was further divided into a non-septic group (47 newborns) and a septic group (34 neonates). Patients with sepsis had considerably lower platelet counts than those without sepsis (p < 0.001). There were no discernible changes between the two groups for the remaining CBC parameters and ZPP/H ratio, despite the sepsis group having considerably higher levels of IT ratio, CRP, and ferritin (p = 0.003, p = 0.002, and p < 0.001, respectively) Table 4.

Table 4

Comparing CBC parameters, ferritin, and ZPP/H of the anemic group regarding sepsis at 6 weeks postnatal age

	Anemic group (81)		Test	P
		Sepsis	No sepsis
		N = 47	N = 34
Hb (g/dl)	Mean±SD	11.5±1.8	11.4±1.7	t = 0.83	0.41
	Range	7–11.5	8–13
HCT (%)	Mean±SD	32.8±5.4	34.3±5.4	t = 0.98	0.33
	Range	21.9–40.4	21–39.5
MCHC (g/dl)	Mean±SD	23.8±5.2	23.5±6.7	t = 0.12	0.89
	Range	21.9–30.1	22–32.6
MCV (fL)	Mean±SD	81.5±6.4	82.3±5.8	t = 0.23	0.75
	Range	70.2–93.5	71–92
Platelets (×10³/L)	Mean±SD	198±117	348±146	U = 5.1	<0.001
	Range	40–440	124–708
TLC (×10³/L)	Mean±SD	13.5±8.8	10.5±4.6	U = 0.61	0.58
	Range	4.8–46.2	3.7–35.5
Reticulocytes (%)	Mean±SD	4±1.9	4.7±1.2	U = 0.79	0.49
	Range	3.1–5.8	3.4–6.2
IT Ratio	Mean±SD	0.16±0.1	0.09±0.09	U = 3.1	0.003
	Range	0.04–0.43	0–0.32
CRP (mg/L)	Median	12	6	U = 3.2	0.002
	Range	1–139	1–96
Serum ferritin (ng/ml)	Median	132.4	33.7	U = 9.4	<0.001
	Range	29.1–662.3	16.2–143.2
Iron (μg/dL)	Mean±SD	82.6±48.4	71.3±57.2	t = 0.33	0.711
	Range	65.4–101.9	54.2–154.8
TIBC (μg/dL)	Mean±SD	213.0±71.1	189.3±55.9	t = 0.24	0.786
	Range	89.2–316.3	81.6–287.8
ZPP/H (μmol/mol)	Mean±SD	122±34	113±28	t = 1.8	0.111
	Range	76–184.2	75.3–176.2

The non-anemic group was divided up even further into a septic group (31 neonates) and a non-septic group (53 neonates). Sepsis patients’ platelet counts were considerably lower than those of non-sepsis patients (p < 0.001) While in the sepsis group, the IT ratio, CRP, and ferritin levels were all considerably higher (p < 0.001 for each). Regarding the other CBC parameters and ZPP/H ratio, no appreciable variations were found between the two groups Table 5.

Table 5

Comparing CBC parameters, ferritin, and ZPP/H of the non-anemic group regarding sepsis at 6 weeks postnatal age

	Non-anemic group (84)		Test	P
		Sepsis	No sepsis
		N = 31	N = 53
Hb (g/dl)	Mean±SD	15±1.8	14.8±1.7	t = 0.59	0.54
	Range	12–20	12–20
HCT (%)	Mean±SD	43.8±6.4	44.6±6.4	t = 0.61	0.53
	Range	36.9–57	34–59.5
MCHC (g/dl)	Mean±SD	34.5±5.4	33.5±6.1	t = 0.22	0.75
	Range	30–44	31–44
MCV (fL)	Mean±SD	103.4±4.5	102.3±5.1	t = 0.26	0.70
	Range	91–112	92–120
Platelets (×10³/L)	Mean±SD	183±81	267±99	t = 4.1	<0.001
	Range	40–371	140–397
TLC (×10³/L)	Mean±SD	10.3±3.8	10.5±4.8	t = 0.28	0.69
	Range	4.6–29.7	4.6–25.5
Reticulocytes (%)	Mean±SD	3.7±1.4	3.4±1.7	t = 0.86	0.35
	Range	2.6–5.8	2.1–6.1
IT Ratio	Mean±SD	0.17±0.08	0.1±0.079	U = 2.1	<0.001
	Range	0.04–0.36	0–0.24
CRP (mg/L)	Median	12	3	U = 10.5	<0.001
	Range	1–139	1–96
Serum ferritin (ng/ml)	Median	129.8	58	U = 6.1	<0.001
	Range	29.1–660	15.2–445
Iron (μg/dL)	Mean±SD	75.5±41.2	67.5±36.0	t = 0.29	0.856
	Range	48.3–178.9	39.1–169.6
TIBC (μg/dL)	Mean±SD	167.0±59.1	146.7±24.0	t = 0.52	0.143
	Range	98.9–203	86.8–187.2
ZPP/H (μmol/mol)	Mean±SD	79.4±15.2	74.2±15.8	t = 0.16	0.88
	Range	43.8–115.1	41.7–132.1

Infants were categorized as belonging to one of four groups, based on the frequency of red blood cell transfusion (before transfusion, after 1st transfusion, after 2nd transfusion, and after 3rd transfusion). Following each transfusion of packed red blood cells, hemoglobin, hematocrit, and serum ferritin all significantly increased (p < 0.001 for each). While the ZPP/H ratio declined dramatically (p < 0.001) with each transfusion of packed red blood cells. In anemic individuals, however, there were no discernible differences in platelet or leucocyte count before and after blood transfusion Table 6.

Table 6

Laboratory characteristics of the Packed red blood cells transfused group

		Packed red blood cells transfused group (N = 47)				P
		Before transfusion	After 1st transfusion	After 2nd transfusion	After 3rd transfusion
		N = 47	N = 47	N = 14	N = 9
Hb (g/dl)	Mean±SD	8.1±2.5	10.1±1.4	12.2±2.1	13.2±1.1	<0.001^a,b,c
	Range	7–10.5	9–13.2	10–14.3	11–15.3
HCT (%)	Mean±SD	22.3±8.5	26.3±5.2	32.2±4.9	39.1±4.4	<0.001^a,b,c
	Range	20–30	24–32	29–41	33–44
Platelets (×10³/L)	Mean±SD	237±111	271±142	254±60	240±14	0.15
	Range	86–442	64–464	151–298	227–252
TLC (×10³/L)	Mean±SD	12.6±5.3	13.7±4.5	11.2±3.1	11.2±3.4	0.089
	Range	4.6–24.3	7–22	5–12	7–20
Reticulocytes (%)	Mean±SD	3.3±2.0	2.0±1.2	2.5±1.0	2.3±1.3	0.446
	Range	2.1–6.4	1.9–5.9	1.3–6.1	1.5–5.8
Serum ferritin (ng/ml)	Mean±SD	117±66.1	151.3±123.8	253.3±196.4	361.3±296.4	<0.001^a,b,c
	Range	24.9–243.2	22.4–463.8	158.5–548	183.6–880.3
Iron (μg/dL)	Mean±SD	69.0±44.3	87.5±65.2	83.7±44.9	78±36.4	0.219
	Range	32.9–189.2	42.9–201	51.2–197.1	36.2–198
TIBC (μg/dL)	Mean±SD	188.8±62.4	202.2±95.5	177.3±62.5	160.3±89.6	0.566
	Range	94.3–336.2	87.5–387.1	81.8–297.6	91.3–274.8
ZPP/H ratio(μmol/mol)	Mean±SD	159.1±15.6	138.2±17.1	122.4±12.6	86.1±6.6	<0.001^a,b,c
	Range	32.2–281.8	22.4–244	23.1–208	22.1–125.5

A comparison was done by one-way ANOVA test, a; the significant difference between before and after 1st transfusion, b; the significant difference between after 1st and after 2nd transfusion, and c; the significant difference between after 2nd and after 3rd transfusion. Hb; hemoglobin, HCT; hematocrit, TLC; total leucocytic count, TIBC; total iron binding capacity, ZPP/H ratio; zinc protoporphyrin to hemoglobin ratio. Bold values are significant at p≤0.05.

When serum ferritin and ZPP/H ratio were compared for their ability to identify anemia using ROC curve analysis, ferritin performed poorly, with an AUC of just 0.58 at a cut-off value of <89.3 ng/ml and 56.7% sensitivity, 53.3% specificity, 54.8% PPV, and 55.2% NPV (p = 0.44). While the ZPP/H ratio had a reasonable AUC (0.97), it may significantly identify anemia with 86.6% sensitivity, 90% specificity, 89.7% PPV, and 87.1% NPV (P < 0.001) at a cut-off value >110.2μmol/mol Fig. 1.

Fig. 1

Receiver operator characteristics (ROC) analysis for performance of serum ferritin and ZPP/H ratio to diagnose anemia in preterm infants.

The cutoff level of >151.6μmol/mol for the ZPP/H ratio level provided the greatest predictive values (93.5% PPV and 94.7% NPV) using the ROC curve, with a sensitivity of 96.7%, specificity of 90%, and AUC = 0.948 (p < 0.001) for the necessity for packed red blood cell transfusions. The serum ferritin cutoff level of <119.2 ng/ml, however, had poor predictive values and an AUC (57.3% PPV, 62.5% NPV, and AUC = 0.566), as well as a sensitivity and specificity of 71.1% and 47.4%, respectively (p = 0.15) Fig. 2.

Fig. 2

Receiver operator characteristics (ROC) analysis for performance of serum ferritin and ZPP/H ratio to predict the need for packed red blood cell transfusion in anemic preterm infants.

4 Discussion

This study aims to investigate the significance of ferritin and zinc protoporphyrin to heme ratio (ZnPP/H) as indicators of iron status in preterm newborns and to ascertain how certain inflammatory events can influence these measurements. A lot of research has been done on the biomarkers for iron status in preterm newborns [3 , 12], It is unclear, though, which iron sufficiency marker in preterm newborns best connects with inflammatory outcomes. Novel markers like ZnPP/H have potential advantages even though more established markers of iron sufficiency, such as ferritin, have been extensively discussed in clinical practice and research literature. Because ZnPP/H is less impacted by inflammation than ferritin, we reasoned that it would be a better marker for determining the iron sufficiency status in preterm newborns. In the current study, the median serum ferritin level and the mean ZPP/H ratio on 1st post-natal day were 51.4 ng/ml (range: 16.8–183.7) and 84.6±18.4μmol/mol, respectively, and were significantly lower than the levels at 6 weeks postnatal age of 87.3 ng/ml (range: 26.2–246.3) and 114.2±26.8μmol/mol, respectively (p < 0.001 for each). This was supported by Carissa et al. [18] ‘s findings that ZnPP/H was greater in the first week of life (120 -53 vs 107 - 36, P = 0.01) when cord blood and the first week of life ZnPP/H levels were compared from the same infants. Whether the discrepancy between cord blood and postnatal levels reflects actual changes that take place during the first few weeks after birth is unknown. The absence of placental transfer of iron and phlebotomy losses, which are highest in the first two postnatal weeks, are factors supporting a true drop in iron stores [19]. However, Griffin et al. [20] showed that plasma ferritin was measured at the same time as ZPP/H is at odds with our findings. 86 patients provided a total of 357 samples. Each infant underwent between 0 and 10 measurements (median four). With increasing postnatal age, plasma ferritin decreased. Less than 20μg/l was the value in 96 cases (27%), less than 15μg/l in 58 cases (16%), and less than 10μg/l in 25 cases (7%). This variation may be explained by the fact that ferritin, an acute phase reactant, is not a great indicator of iron storage because it can falsely increase under infectious settings, and there is very weak evidence that it is useful in assessing iron stores in preterm newborns [20, 21].

The median serum ferritin level at 6 weeks postnatal age was found to be significantly lower in anemic neonates (44.5 ng/ml with a range of 22.3–165) than it was in non-anemic neonates (73.6 ng/ml with a range of 16.8–248.3) (p = 0.044), While the mean serum ZPP/H level in anemic neonates was substantially higher (115.2±33.4μmol/mol) than the mean serum level in non-anemic neonates (85.6±21.3μmol/mol)(p < 0.001). This was in line with other studies that demonstrated how ferritin, the body’s main form of iron storage, is typically low when there is an iron deficiency. Poorer neurodevelopmental outcomes have been linked to ferritin levels below the standard cutoff of 12 ng/mL [22 –25]. On the other hand, when there is an iron deficiency, zinc is added to the protoporphyrin ring in place of iron to produce heme. When there is less iron available, the ratio of zinc to iron, or ZnPP/H, is higher [25]. According to the current study, the median level of serum ferritin in patients with sepsis was considerably higher than the median level in patients without sepsis (124 ng/ml with a range of 16.8–662.3) (75 ng/ml with a range of 16.2–316.1) (p < 0.001), While the means of ZnPP/H (114.3±26.5 and 123.8±28.5μmol/mol), did not significantly differ between septic and non-septic patients (p = 0.062). Additionally, the median of serum ferritin in both anemic and non-anemic septic patients (132.4 ng/ml with a range: of 29.1–662.3 and 129.8 ng/ml with a range: of 29.1–660) was significantly higher than its median in both anemic and non-anemic non-septic patients (33.7 ng/ml with a range: 16.2–143.2 and 58 ng/ml with a range: 15.2–445) (p < 0.001each). However, the means of ZnPP/H of anemic and non-anemic septic patients (122±34 and 79.4±15.2μmol/mol) did not show a significant difference when compared to the means of anemic and non-anemic non-septic patients (113±28 and 74.2±15.8μmol/mol) (p = 0.11 and p = 0.88). Our results are in line with those of German et al., who stated that ferritin was greatly impacted by clinical inflammatory conditions, such as sepsis with a positive culture. This is in line with ferritin’s established to function as an acute phase reactant. The mean log values of the ferritin changed significantly during sepsis, with an adjusted difference of 1.19 ng/mL (adjusted p < 0.001) between them. Infants having a positive blood culture within 2 days of or 1 day after ZnPP/H assessment showed no discernible difference in ZnPP/H values. Since ZnPP/H detects changes in heme synthesis and erythrocyte turnover, it is plausible that ZnPP/H represents a more stable indicator of iron status during periods of inflammation or that the effects of inflammation on ZnPP/H are delayed [16]. Our findings agree with Griffin et al. who revealed that ferritin is not a precise measure of iron stores and may be falsely elevated under infectious conditions and that there is only indirect evidence of its value as a measure of iron stores in premature newborns. Low values, however, are considered to be strongly suggestive of iron insufficiency, and it continues to be the most commonly used indicator of iron status [20]. When serum ferritin and ZPP/H ratio were compared for their ability to detect anemia using ROC curve analysis, ferritin had a poor AUC (0.582) at a cut-off value of <8 9.3 ng/ml, with 56.7% sensitivity and 53.3% specificity (p = 0.44). While the ZPP/H ratio had a good AUC (0.97), it can significantly identify anemia at a cut-off value >110.2μmol/mol with 86.6% sensitivity and 90% specificity (p < 0.001). Additionally, utilizing the ROC curve, the cutoff level of >151.6μmol/mol for the ZPP/H ratio level offered the greatest predictive values (93.5% PPV and 94.7% NPV), with a sensitivity of 96.7%, specificity of 90%, and AUC = 0.948, for the necessity for packed red blood cell transfusions (p < 0.001). While, the cutoff level of <119.2 ng/ml For serum ferritin showed poor predictive values and AUC (57.3% PPV, 62.5% NPV, and AUC = 0.566) with a sensitivity of 71.1%, and specificity of 47.4% (p = 0.15). This was in harmony with Serdar et al. study which stated that ZPP had the highest sensitivity for detecting children with iron deficiency anemia (IDA) [26]. Furthermore, according to receiver operating characteristic (ROC) curves, measures of hemoglobin had a limited sensitivity (23–40%) for predicting iron deficiency (ID) [27] and lacked sensitivity (11%) among preschool-aged children in a study by Crowell et al. [28]. Similarly, Yu et al. reported that the ZPP/H ratio was a more sensitive marker to diagnose iron deficiency than hemoglobin or serum ferritin (SF) measurement, properly identifying more than twice as many children (sensitivity of 91.7% for ZPP/H ratio, compared to 41.7% for hemoglobin or SF). In other words, compared to hemoglobin or SF measures, the ZPP/H ratio alone as a screening test for ID only missed 8.3% of the children who were iron deficient. As a result, it becomes essential to incorporate the ZPP/H ratio into future screening regimens for the early identification of ID in newborns and young children. However, compared to hemoglobin or SF, the ZPP/H ratio had lower specificity (60.2% vs. 89.1% or 96.4%, respectively) and led to the false identification of more subjects who weren’t genuinely ID. Only 1 out of every 10 children with a normal iron status test had a false-positive rate with the defined hemoglobin cutoff value, compared to 40% for the ZPP/H ratio (4 of every 10 normal children tested) [29]. These cutoff values could be unique to our research group and could differ in other circumstances or patient populations. Before applying them in clinical practice, it is crucial to validate these cutoff values in additional cohorts.

According to our study, there was a significant rise in serum ferritin following each transfusion of packed red blood cells (p < 0.001 each). While the ZPP/H ratio considerably dropped after each transfusion of packed red blood cells (p < 0.001 each). Our findings supported those of a prior study by German et al., who found that red blood cell transfusions given within 7 days of the laboratory measurement significantly affected ferritin concentrations, with an adjusted difference in mean log values of 1.03 ng/mL (adjusted P.001) [16]. Red blood cell (RBC) transfusion has been shown to alter the ZnPP/H in two different ways, according to Winzerling et al. First, as the transfused cells degrade and the iron is recycled, the infant can utilize the iron as a source from the transfused cells. Second, the ZnPP/H produced by the infusion of adult RBCs is a mixture of the infant’s iron status and the iron status of the adult who donated the cells. 15% –18% of the circulating RBCs are supplied by a transfusion of 15 mL/kg. For this reason, one would anticipate an initial drop in ZnPP/H following transfusion, given that the typical adult ZnPP/H ranges from 30 to 80 [30]. However, Dani et al. demonstrated that the transfusion-mediated stimulation of the inflammatory cascade may be consistent with the immediate increase in ferritin upon transfusion [31]. Rao et al., however, observed that the fact that preterm newborns have a large quantity of iron in the red cell mass and smaller amounts in the tissues as storage iron (ferritin) could account for unchanging ferritin levels before and after transfusion. When iron supply and demand are not balanced in babies, iron is prioritized for Hb production in the RBC [32].

5 Conclusion

It is evident from our depicted results that iron sufficiency can be assessed by serum ferritin or ZnPP/H assays; however, ZnPP/H may be more reliable as a marker of iron status during particular inflammatory events, such as infection and packed red blood cell transfusion. Further studies are needed to clarify the sensitivity and specificity of serum ferritin or ZnPP/H concerning inflammation in preterm and term infants.

Conflict of interest

The authors declare no conflict of interest.

Footnotes

Acknowledgments

We sincerely thank all the subjects who participated in this study and the clinicians who helped in recruiting the preterm infants.

Funding

Authors didn’t receive any source(s) of support in the form of grants.

Ethics approval and consent to participate

The current study was approved by the Medical Research Ethical Committee of the Faculty of Medicine, Benha University. All subjects were informed about the procedures and the aim of the study and informed written consent was obtained from the parents or caregivers of enrolled infants. The committee’s reference number is Ms14-12-2020.

References

Angulo-Barroso

, Li

, Santos

DCC

, Bian

, Sturza

, Jiang

, et al. Iron supplementation in pregnancy or infancy and motor development: A randomized controlled trial, Pediatrics 2016;137(4).

Fretham

SJB

, Carlson

, Wobken

, Tran

P V

, Petryk

, Georgieff

Temporal manipulation of transferrin-receptor-1-dependent iron uptake identifies a sensitive period in mouse hippocampal neurodevelopment, Hippocampus 2012;22(8):1691–702.

Lukowski

, Koss

, Burden

, Jonides

, Nelson

, Kaciroti

, et al. Iron deficiency in infancy and neurocognitive functioning at 19 years: Evidence of long-term deficits in executive function and recognition memory, Nutr Neurosci 2010;13(2):54–70.

Widness

, Madan

, Grindeanu

, Zimmerman

, Wong

, Stevenson

Reduction in red blood cell transfusions among preterm infants: Results of a randomized trial with an in-line blood gas and chemistry monitor, Pediatrics 2005;115(5):1299–306.

Buonocore

, Perrone

, Bracci

Free radicals and brain damage in the newborn, Biol Neonate 2001;79(3–4):180–6.

Baker

, Greer

Diagnosis and prevention of iron deficiency and iron-deficiency anemia in infants and young children (0-3 years of age), Pediatrics 2010;126(5):1040–50.

Agostoni

, Buonocore

, Carnielli

, De Curtis

, Darmaun

, Decsi

, et al. Enteral nutrient supply for preterm infants: Commentary from the European society of paediatric gastroenterology, hepatology and nutrition committee on nutrition, J Pediatr Gastroenterol Nutr 2010;50(1):85–91.

Fretham

SJB

, Carlson

, Georgieff

The role of iron in learning and memory, Adv Nutr 2011;2(2):112–21.

Cook

, Baynes

, Skikne

Iron deficiency and the measurement of iron status, Nutr Res Rev 1992;5(1):198–202.

10.

Labbé

, Dewanji

Iron assessment tests: Transferrin receptor vis-à-vis zinc protoporphyrin, Clin Biochem 2004;37(3):165–74.

11.

National Committee for Clinical Laboratory Standards. C42-A erythrocyte protoporphyrin testing; Approved Guideline. 1996;(November).

12.

Rettmer

, Carlson

, Origenes

, MD

, Jack

, Labbé

Zinc protoporphyrin/heme ratio for diagnosis of preanemic iron deficiency, Pediatrics [Internet] 1999;104(3):e37–e37.

13.

Puri

, Suresh

, Gogtay

, Thatte

Declaration of Helsinki, Implications for stakeholders in research, J Postgrad Med 2009;55(2):131–4.

14.

Jopling

, Henry

, Wiedmeier

, Christensen

Reference ranges for hematocrit and blood hemoglobin concentration during the neonatal period: Data from a multihospital health care system, Pediatrics 2009;123(2):e333–7.

15.

Christensen

, Jopling

, Henry

, Wiedmeier

The erythrocyte indices of neonates, defined using data from over 12,000 patients in a multihospital health care system, J Perinatol Off J Calif Perinat Assoc 2008;28(1):24–8.

16.

German

, Vu

, Grelli

, Denton

, Lee

, Juul

Zinc Protoporphyrin-to-heme ratio and ferritin as measures of iron sufficiency in the neonatal intensive care unit, J Pediatr 2018;194,47–53.

17.

Walsh

, Kliegman

Necrotizing enterocolitis: Treatment based on staging criteria, Pediatr Clin North Am 1986;33(1):179–201.

18.

Cheng

, Zerzan

, Johnson

, Juul

Zinc protoporphyrin-to-heme ratios in high-risk and preterm infants, J Pediatr 2012;161(1):81–7.e1.

19.

Valieva

, Strandjord

, Mayock

, Juul

Effects of transfusions in extremely low birth weight infants: A retrospective study, J Pediatr 2009;155(3):331–37.e1.

20.

Griffin

, Reid

, McCormick

KPB

, Cooke

Zinc protoporphyrin/haem ratio and plasma ferritin in preterm infants, Arch Dis Child Fetal Neonatal Ed 2002;87(1):F49–51.

21.

Thomas

Nutritional care of preterm infants: Scientific basis and practical guidelines, Indian J Med Res 2016;143,531–2.

22.

Lozoff

, Jimenez

, Hagen

, Mollen

, Wolf

Poorer behavioral and developmental outcome more than 10 years after treatment for iron deficiency in infancy, Pediatrics [Internet] 2000;105(4):E51.

23.

Shafir

, Angulo-Barroso

, Jing

, Angelilli

, Jacobson

, Lozoff

Iron deficiency and infant motor development, Early Hum Dev 2008;84(7):479–85.

24.

Carter

, Jacobson

, Burden

, Armony-Sivan

, Dodge

, Angelilli

, et al. Iron deficiency anemia and cognitive function in infancy, Pediatrics 2010;126(2):e427–34.

25.

German

, Vu

, Neches

, Juul

Comparison of two markers of iron sufficiency and neurodevelopmental outcomes, Early Hum Dev 2021;158,105395.

26.

Serdar

, Sarici

, Kurt

, Alpay

, Okutan

, Kurnaz

, et al. The role of erythrocyte protoporphyrin in the diagnosis of iron deficiency anemia of children, J Trop Pediatr 2000;46(6):323–6.

27.

Schneider

, Fujii

, Lamp

, Lönnerdal

, Dewey

, Zidenberg-Cherr

Anemia, iron deficiency, and iron deficiency anemia in 12-36-mo-old children from low-income families, Am J Clin Nutr 2005;82(6):1269–75.

28.

Crowell

, Ferris

, Wood

, Joyce

, Slivka

Comparative effectiveness of zinc protoporphyrin and hemoglobin concentrations in identifying iron deficiency in a group of low-income, preschool-aged children: Practical implications of recent illness, Pediatrics 2006;118(1):224–32.

29.

Effectiveness of zinc protoporphyrin/heme ratio for screening iron deficiency in preschool-aged children, Nutr Res Pract 2011;5(1):40–5.

30.

Winzerling

, Kling

Iron-deficient erythropoiesis in premature infants measured by blood zinc protoporphyrin/heme, J Pediatr 2001;139(1):134–6.

31.

Dani

, Poggi

, Gozzini

, Leonardi

, Sereni

, Abbate

, et al. Red blood cell transfusions can induce proinflammatory cytokines in preterm infants, Transfusion 2017;57(5):1304–10.

32.

Rao

, Georgieff

Perinatal aspects of iron metabolism, Acta Paediatr Suppl 2002;91(438):124–9.