Association of Birth Weight with Gestational Age and Maternal Measles Antibodies

Address for correspondence: Dr. Baba Usman Ahmadu Head of Department, Paediatrics, Federal Medical Centre Yola, P.M.B 2017, Yola, Adamawa State, Nigeria E-mail: ahmadu4u2003@yahoo.com Phone: +2348033668948 1Dr. Baba Usman Ahmadu (MBBS, MHPM, FMC Paed), Head of Department, Paediatrics, Federal Medical Centre Yola, P.M.B 2017, Yola, Adamawa State, Nigeria. Formerly of the Department of Paediatrics University of Maiduguri Teaching Hospital. Visiting consultant Paediatrician Abubakar Tafawa Balewa University Teaching Hospital Bauchi and Federal Medical Centre Jalingo, Taraba state, Nigeria, 2Yohanna Shemiya (MBBS), 3Godfrey Anita (MBBS), 4Shammah Ubong (MBBS), 5Bello Abdulrahman (MBBS), 6Bello Amina (MBBS), 7Yanshak Iranyang (MBBS), 8Zabadi Rabi (MBBS), 9Madu Anselem Chekwube (MBBS), 10Mana Umar Fatima (MBBS), 11Auta Gidado (MBBS) All from the Department of Paediatrics, Federal Medical Centre, Yola, Adamawa state, Nigeria. Abstract


Introduction
M easles is the principal cause of vaccine-preventable deaths in infants in the world.The Global burden of disease (GBD) project in 2000, es mated that of 1.7million deaths associated with vaccine preventable disease, measles accounted for about 46% 1,2 .Several countries in Africa, South-East Asia, Europe, Eastern Mediterranean and Western Pacifi c regions recorded the resurgence of measles whenever measles control prac ces were relaxed 3 .Cases of measles were reported across con nents and countries of the world like Cameroon, Bangladesh and the United States of America to men on just a few 3,4,5,6   .Maternal measles an body (MMA) protect infants from measles and are transferred to foetus from third trimester of gesta on, the age at which the weight of the foetus also increases signifi cantly under favourable in utero environment 7,8,9 .
Varia on in the prevalence of MMA in babies has been found to be dependent on geographical and socio-demographic factors 1,3,10,11 .Several factors were studied with a view to fi nd reasons for varia on within popula ons.Well-documented ones are prior measles virus (MV) exposure that produces higher MMA than measles immuniza on.Previous study conducted by Bromberg et al 12 in 1994, demonstrated that mothers from developing na ons had higher levels of MMA than those in developed countries.Another study carried out in the United Kingdom by Brugha et al 13 in 1996, similarly showed a higher MMA in mothers with a history of natural MV exposure.Other variables that had been studied included socioeconomic status, educa on, race, parity, age, nutri on and gesta onal age (GA).All these apart from GA looked at MMA of mothers as a surrogate determinant of measles an bodies in babies 8,9,10 .
Gesta onal age was found to be associated with diff erent levels of MMA in babies such that preterm babies may have lower levels than their term or pos erm counterparts 8,9,10 .Researchers in Sri Lanka and some part of Africa had looked at the infl uence of birthweight (BW) on MMA of babies; however, the subject ma er is s ll not very clear due to paucity of literatures [14][15][16] .Moreover, studies showing the combine eff ects of BW and GA on MMA are almost lacking.Learning more about the associa on between BW and its combined eff ect with GA on the star ng levels of MMA in babies [14][15][16] , cannot be overemphasized because of the signifi cance it could have in protec ng babies against measles later in infancy and early childhood.Therefore, the aims of this work were four :-1) to determine the associa on of BW with GA, 2) to assess the rela onship between BW and MMA, 3) to correlate GA with MMA and 4) to examine the combine eff ects of BW, GA on MMA.To the best of our knowledge there are no studies or data on this subject ma er especially in Maiduguri, North-Eastern Nigeria; and paucity of informa on s ll exist on this issue especially in developing countries of the world a er vast literature review.

Material and Methods
The study was conducted at the Department of Paediatrics, Immunology and Obstetrics unit of the University of Maiduguri Teaching Hospital (UMTH), Nigeria.The UMTH is a ter ary centre located in North-Eastern Nigeria and a centre of excellence in infec ous diseases and immunology.The study area is defi ned by longitude 13.50 E and la tude 110 N. The characteris c vegeta on is that of Sub-sudan and Guinea savannah with an average annual rainfall of 79mm and 197mm in the northern and southern part of the state.Temperature ranges from 15 •C to 39.7 •C.Being the largest health facility in the region, the UMTH serves as a referral site for the six North-Eastern States and neighboring countries of Chad, Cameroon and Niger Republics.The study was a hospital-based cross-sec onal descrip ve study of babies recruited from the labour ward of the UMTH.
Par cipa on in this study was voluntary and consen ng mothers were selected using systema c random sampling method where the fi rst of every fi ve mother was picked as they presented to the labour ward, using the labour ward register as a sampling frame.Where the fi rst did not fulfi ll the inclusion criteria, the immediate next mother that qualifi ed was selected.Pregnant women who delivered at the UMTH and consented to the study had their babies enrolled in this work.Sick babies or those whose mothers decline consent were excluded.
The study protocol was reviewed and authorised by the Medical Research and Ethics Commi ee of UMTH in compliance with guidelines of the Helsinki declara on on biomedical research on human subjects.Assistance of linguis cs interpreters of informed consent form in local languages mainly (Kanuri and Babur) were sought for due to low literacy rate in Maiduguri 17 .Parents had unlimited liberty to deny consent without any consequences while confi den ality was maintained.

Sample Size and collec on of specimens:
The minimum sample size was determined using a sta s cal formula 18 , which compares means and standard devia on based on Jacob Cohen eff ect size of 0.2, alpha levels of 0.05 and power of 90%.This equalled 130; however, 50% of the sample size was added to maximize power.Thus the total sample size for this study was 200 babies.
The BW of babies was measured using the bassinet weighing scale that has a sensi vity of 50gms set at zero mark.Birthweight > 3.99 kilograms (kg) was classifi ed as macrosomia, 2.5 -3.99 (kg) was normal, 1.5 -< 2.5 (kg) was low BW, 1.0 -< 1.5 (kg) was very low BW and < 1.0 (kg) was extremely low BW 19 .Gesta onal age of babies was determined using the mothers' last menstrual period or by Dubowitz score or by Obstetric ultrasound scan where available 20 .
Two millilitres (mls) of neonatal cord blood were obtained at birth using sterile disposable fi ve mls syringe under asep c technique, and placed in sterile plain bo les.Sera were separated a er centrifuging these blood samples at 5000 revolu ons per minute (rpm) for fi ve minutes.The sera were used to es mate MMA (U/ml) by enzyme linked immunosorbent assay (ELISA).All sera collected were pooled in a refrigerator at -20°C un l the me of MMA assay.Data analyses were performed using SPSS sta s cal so ware version 16, Illinois, Chicago USA and a computer program for epidemiologist PEPI version 3.01.Descrip ve data were presented in Tables and Figures were used for illustra on.Spearman's correla on (rho) for BW and MMA and Kendall's rank correla on (tau b) for GA and MMA were determined.Jonckheere-Terpstra test of associa on between BW and MMA of babies was also inves gated.Univariate analysis was used to inves gate the combine eff ects of BW and GA on MMA.Likely-hood ra o Chi-square (χ2) was used to compare categorical variables.Goodman-Kruskal index rank order of predic ve associa on (tau) for BW and GA of babies was also determined.A p value < 0.05 was considered signifi cant.

Results
Two hundred babies were enrolled in this study.There were 101 (50.5%) males and 99 (49.5%)females.The male to female ra o was 1.02:1.Majority of the babies 169 (84.5 %), 117 (58.5%) were within the range for normal BW and term respec vely (Table 1).The babies mean BW was 3.04 (0.58) at 95% CI (2.96 -3.12) kg.Table 2 shows BW and GA distribu on of the subjects.Associa on between BW and GA was signifi cant (p=0.003).Overall, the Goodman-Kruskal index of predic ve associa on (tau) for BW and GA was signifi cant (p=0.039); that for predic ve associa on of variable classifi ed in the rows (BW) was (p =0.012) and columns (GA) was (p=0.052)respec vely.
Table 3 shows BW, GA, median MMA and confi dence intervals.Figure 1 and 2 below depicts pictorial area of rela onship between BW, MMA and GA.Speaman's correla on (rho) for BW and MMA was 0.400 but Jonckheere-Terpstra test of associa on between these variables was not signifi cant (p=0.087).Though Kendall's rank correla on (tau b) for GA and MMA was signifi cant (p<0.001), the combine eff ects of BW and GA on MMA using univariate analysis was not signifi cant (p=0.587).

Discussion
Majority of babies in this study had their BW within the normal range and were delivered at term.This may not be unconnected to increased awareness and patronage of antenatal care (ANC) services by women during pregnancy.This view agreed with that of researchers at Sylhet, Omani Medical College Hospital in 2010 5 .Nigeria's country health system record has shown that up to 61 percent of pregnant women had at least one ANC visit and about half of these women had skilled a endant receiving their babies at the me of delivery 21 .Both preven ve and interven on ac vi es with regard to maternal health are conducted during ANC that could have led to healthy women carrying healthy adequate weight babies to term.
Comparing BW of babies with MMA was not signifi cant in current study.Similar observa on was made by other workers where they reported that, MV exposure and measles immuniza ons as chief determinants of MMA could mask other infl uencers of measles an body like BW 8 .In support of this, some authors associated measles with lifelong an bodies and others argued that even measles immuniza on gives an bodies that could last up to four decades, and possibly lifelong 8,9,17,22 .Moreover, our study popula on was found in measles endemic region, which further bu resses MV exposure and measles immuniza on as principal determinant of MMA 8,17,21 .The former through boos ng eff ect of MV and the la er through measles an body mediated passive humoral immunity.During the third trimester of pregnancy, MMA are ac vely passed in mother-foetal pairs 8,9,10 .This possibly explains the levels of MMA observed in VLBW, LBW and advancing GA of babies that make up our study popula on.The rela onship between GA and MMA has been well published [8][9][10] .Term and posterm babies usually have higher MMA because of the recruitment and upward regula on of MMA receptors associated with foetal maturity.Gain in MMA receptors is known to ferry more of these an body across the placenta to the foetus.
Other inves gators, however, had found that BW of babies was directly propor onal with MMA 23 .Foetal adapta on mechanisms could be the probable reason for their fi ndings because more placentomes are recruited as the weight of the foetus increased resul ng to increased placental volume ra o 24 .This would enable increased transplacental transfer of MMA in mother-foetal pair.However, several other colleagues have debated on the issue of foetal weight versus placental size correla on as a determinant of substrate transfer to the foetus 25 .They believed that substrate transfer to the foetus is not just a serial but a programmed parallel and complex developmental event of the foetus.Therefore, there is the need for further research in this aspect.
Our work has shown that BW was linearly related with GA of babies; however, no signifi cant observa on was made between MMA and BW.Similarly, the combine eff ect of BW and GA yielded no signifi cant eff ect on MMA in present work.From this fi nding, it can be deduced that GA rather than BW may be a be er determinant of MMA.Since BW and its combined eff ect with GA did not contribute signifi cantly towards MMA genera on, the principal determinants of measles an body in babies may rest with their mothers.As such these babies may need measles immuniza on because their MMA levels could be as a result of measles campaign as was seen in Sabongidda-Ora, Edo State, and other parts of Nigeria and also other measles endemic sites 8,26 .Measles an bodies due to measles campaign rapidly decay leaving these babies suscep ble to measles early in infancy 8 .Therefore, more health facility should be modeled in such a way that measles immuniza on can be administered to children in addi on to mass campaigns.

Conclusions
Birthweights vary directly with GA and MMA was found to have a linear rela onship with GA; however, no signifi cant rela onship was established for BW alone or in combina on with GA on MMA.

Limitations
This work was conducted in a single hospital (UMTH); therefore, there is the need to be cau ous in generalizing the data of this study.More so, the study did not look at MMA that was associated with maternal measles immuniza on or prior wild type MV infec on of mothers.This was because of lack of facility in our centre that could diff eren ate these an bodies one from another.More so, most of these mothers were un-able to remember whether they had measles immuniza on or wild type MV infec on during childhood.

Recommendations
Based on the fi ndings of this work, it is recommended that mul ple centers' should be involved in future research of this nature to enable generaliza on of fi ndings.Also, future work should address co-founders like maternal measles immuniza on and maternal prior wild type MV infec on.

Table 1 :
Showing birthweight and gesta onal age characteris cs of 200 babies

Table 2 :
Showing birth weight and gesta onal age distribu on of the study cohorts

Table 3 :
Showing birth weight, gesta onal age and maternal measles an body profi les of the babies