Prevalence of Anaemia in Children Diagnosed with Pneumonia in a Tertiary Hospital in Quito , Ecuador

1David Garrido, Servicio de Oncohematología, Hospital Pediátrico Baca Ortiz, Quito, Ecuador, 2Michelle Fuseau, Posgrado de Reumatología, Universidad de la Republica, Montevideo, Uruguay, 3Santiago Garrido, Unidad de Cuidados Intensivos Instituto Ecuatoriano de Seguridad Social, Hospital General Ibarra, Ibarra, Ecuador, 4Gina Vivas, Servicio de Pediatría, Hospital de Especialidades de las Fuerzas Armadas N°1, Quito, Ecuador, 5Miguel Gutiérrez, Servicio de Pediatría, Hospital de Especialidades de las Fuerzas Armadas N°1, Quito, Ecuador. Abstract


Introduction
W ith approximately 1.62 billion (25%) people of the global population aff ected, anaemia is one of the most frequent diseases, primarily aff ecting people from developing or low-income countries.Furthermore, anaemia had become a signifi cant public health problem that aff ects all ages of the world's population, with its highest prevalence among children under fi ve years of age and pregnant women 1 .Anaemia is characterized by a wide variety of etiological factors.Some of these factors include the inadequate nutritional intake of micronutrients (vitamin B12, folate, and iron defi ciencies), non-transmissible disorders, infectious diseases (HIV/AIDS, tuberculosis, malaria, helminths infections, and general infl ammatory conditions), socioeconomic factors, and demographic circumstances, genetic and immuno-haematological disorders 2 .According to the World Health Organization (WHO), 50% of an estimate of 293 million young children and 468 million non-pregnant women are aff ected by anaemia caused by iron defi ciency; mostly due to inadequate iron intake (3) .In Ecuador, there are few studies which report the frequency of anaemia in children; however, it varies between 12% and 45%. 4 .As relevant as anaemia, malnutrition also aff ects our paediatric population.According to UNICEF, chronic malnutrition aff ects 25.3% of the Ecuadorian children 5 .In addition to this report, during the ninth epidemiological week of 2018 in Ecuador (From February 25th to March 3rd), 477 cases of acute malnutrition have been reported, of which 77.98% (372) cases correspond to moderate acute malnutrition, and 22.01% (105) cases correspond to severe acute malnutrition 6 .
Iron functions in the immune system are essential, among which are included: the maintenance of the activity of phagocyte oxidase; the increase of expression of the inducible nitric oxide synthase (iNOS).Similarly, intracellular iron promotes the activation of factor nuclear kappa B (NF-κB) and toll-like receptor 4 (TLR4) signaling 7 .Therefore, there is an association between iron defi ciency and a reduced immune response to infection.On the other hand, an iron excess can be used by bacteria to improve its growth 8 .This association between iron and the immunological system could explain an increased risk of respiratory tract infection observed in children with iron defi ciency anemia 9 .Pneumonia is a frequent disease in children; and with an approximate 17% of 6.3 million child deaths in 2010, was the leading cause of death in children in that year.Moreover, the pneumonia global incidence in children under fi ve years old from low-income and middle-income countries was recently estimated as 0.22 episodes per child-year, of which 11.5% met criteria for severe disease 10 .In Latin America and the Caribbean, community-acquired pneumonia is a substantial cause of morbidity and mortality among children, with an estimated of 327,000 cases and 12,000-28,000 deaths annually 11 .This study aims to present the prevalence of anaemia and malnutrition among Ecuadorian children diagnosed with pneumonia, hospitalized in a tertiary hospital in Quito, Ecuador and to compare whether anaemia could have an impact on the length of hospital stay in pneumonic children.

Material and Methods
This observational and cross-sectional descriptive study was conducted at the Paediatrics Department of the Hospital of Specialties of the Armed Forces N ° 1 (Hospital de Especialidades de las Fuerzas Armadas No. 1 [HE-1]), from September 2017 to December 2017.
All the paediatric patients diagnosed with pneumonia, between July 2016 and September 2017, and its respective clinical history number using the logbooks from the Paediatrics Department.Through the software from the HE-1, the data recorded in the clinical history of each patient was extracted, which include the detailed clinical evaluation and all the complementary tests.
From the analysis of the HE-1 database, a total of 80 children, between six months up to 15 years of age were included.From a potential group of 100 patients, 20 children were excluded.The exclusion criteria for this study were: age outside the interval established for this work, lack of registered anthropometric parameters, absence of a complete blood count (CBC) at the admission to the hospital, diagnosis of concomitant conditions that could aff ect anthropometric or haemoglobin parameters, or that could predispose to pneumonia.No sample size determination was done, as all the children diagnosed with pneumonia were considered.As the anthropometric measurements and hematological parameters determination were perceived as part of the clinical evaluation of each patient, no experimentation was done during the clinical examination.Also, all the procedures were accomplished according to national and international guidelines, to arrive to an accurate diagnosis.Thus, there were no risks to the patients involved in this study.

Pneumonia diagnosis:
The physical fi ndings such as: fever, tachypnoea, breathing diffi culties, rhonchi, crackles, and wheezing were used to establish a pneumonia diagnosis.Then, the evaluation of the hydration status, activity level, and oxygen saturation were essential for the decision of hospitalization and treatment 12 .According to World Health Organization (WHO), the tachypnea thresholds to diagnose pneumonia are: in children between two and 12 months, 50 breaths per minute (BPM) (Normal 25-40 BPM); 40 breaths for minute BPM in children between one and fi ve years (Normal 20-30 BPM); and more than 20 BPM in children over fi ve year's age 13 .Hypoxemia, defi ned as a sustained saturation of peripheral oxygen (SpO2) <90 %, was used as criteria for hospitalization, along with criteria for respiratory distress, which includes: tachypnoea, dyspnoea, retractions (suprasternal, intercostal, or subcostal), grunting, nasal fl aring, apnoea and altered mental status.Furthermore, CBC, acute-phase reactants and chest radiography were performed 14 .

Body mass index:
In determining the weight and height of the children involved in this work, was used the Seca 700 Physician's Balance Beam Scale with height rod and a minimum measurement of 50g (Hospital code: HG-1-03646).Additionally, we used the Digital Baby Scale RL-DBS (Capacity: 20 kg x 0.01 kg, Platform Dimensions: 59.69 x 27.94 cm).The BMI was calculated using the Quetelet index (BMI = weight [Kg]/ height 2 [m 2 ]).After obtaining the BMI, all the children were classifi ed according to the nutrition guidelines for the primary prevention and control of overweight and obesity in children and adolescents established by the Ministry of Public Health (Ministerio de Salud Pública [MSP]), which coincides with the WHO parameters 15,16 .Using the Z-score and percentile classifi cation, the weight, height, and BMI were assessed for all the children included in this study.(16).Under two years of age, we used the weight-for-length charts.

Determination of haematological parameters:
Haematological parameters were determined through an auto haematology analyser Mindray BC-6800, which uses a 3D analysis obtaining information from scattering of laser light at two angles and fl uorescence signals, as well as a colorimetric method to measure haemoglobin.

Anaemia diagnosis:
To diagnose anaemia, we based our criteria on the haemoglobin values suggested by the WHO, when the levels were below the normal for age.Additionally, we used haemoglobin correction according to the geographical residence, as advised by the WHO (17) .We also included in the analysis other haemato-metric parameters presented by the CBC, such as haematocrit (HCT), mean corpuscular volume (MCV), mean corpuscular haemoglobin (MCH), and red cell distribution width (RDW).The corresponding normal values considered in this study are shown in Tables 1  and 2 18 .
In the present study, the exposure factor was anaemia or early markers of iron defi ciency such as high RDW or low MCV.Therefore, two patient groups were included in the study, one of which was anaemic patients diagnosed with pneumonia, and the other was non-anaemic patients diagnosed with pneumonia.There were variables beyond the control of this study, such as ethnic diff erences, iron intake and poverty status.No cases of cigarette usage or pregnancy were founded.
The clinical data were analysed using SPSS software and Excel, both in their latest versions for Windows 10.The risk association between anaemia and pneumonia was calculated through the odds ratio (OR).To contrast the length of hospital stay among de diff erent conditions studied we used the Mann-Whitney U Test.Also, we used convenience sampling.

Results
From a total of 80 patients, we found 47.50% of them being under three years of age.In spite of we classifi ed the patients into two groups according to the age (minors and over three years), we observed that 29 were under 24 months, 19 between two and three years, 18 between four and six years, and the remaining ones were between seven and 15 years.Besides, there was an average age of 3.98 years (SD 3.35; CI 3.25-4.72).
In the whole group of participants, the percentage of female children was higher than males (51.25% vs. 48.75%,respectively).In addition to this, we observed that 31.25% of the participants suff ered a previous episode of pneumonia.
Nutritional status: The most frequent nutritional problem observed in our studied population was low weight (16.25%), being more common in children under three years of age.However, we found anthropometric parameters under Z-score -1 (Weight Z score <-1, 40%; Height Z score <-1, 33.75%; BMI Z score <-1, 30%).Both, severe low weight and severe low height were observed in 3.75% of the patients.There were no children severely emaciated.
Anaemia prevalence: Anaemia was found in 21.25% of the whole group.However, the children under three years of age were more aff ected by this condition (18.75%).Early markers of iron defi ciency, such as mean corpuscular volume and red cells distribution width were more frequent that anaemia itself; 38.75% and 28.75%, respectively.

Risk of anaemia in toddlers:
When the prevalence of anaemia was compared between children under 36 months old (toddlers) and the patients with age over three years, we found a signifi cant risk of anemia in the fi rst group (OR 13.04; CI 95% 11.48-14.61;p<0.05) and low HCT (OR 14.64; CI 95% 12.53-16.75;p<0.05).This pattern was also seen in the early markers of iron defi ciency, low MCV (OR 3.23; CI 95% 2.3-4.17;p<0.05), high RDW (OR 2.77; CI 95% 1.76-3.78;p<0.05), and MCH (OR 8.48; CI 95% 7.14-9.82;p<0.01)Risk association between the anthropometric variables and anaemia: We didn't observe any risk association when the anthropometric variables were considered as exposure factor to suff er anaemia.

Length of hospital stay:
We found no signifi cant diff erence in the length of hospital stay during the comparison of various conditions.Table 4 presents the average number of days that the patients included in this study stayed in HE-1.No cases of death were reported.

Discussion
Anaemia is a disease which is a public health problem in children under six years of age in Latin America and the Caribbean, and is considered a moderate health problem (prevalence of anaemia between 20% and 40%) in Nicaragua, Brazil, Mexico, Ecuador, El Salvador, Cuba, Colombia, Dominican Republic, Peru, Panama, Honduras, and Guatemala 19 .According to the ENSANUT-ECU 2012 (Encuesta Nacional de Salud y Nutrición-Ecuador 2012, National Health and Nutrition Survey-Ecuador 2012), the prevalence of anaemia was 25.7% in children; but, in the indigenous community, percentage reported reached 41.6% 20 .These reports are similar to our fi ndings.However, our anaemia prevalence is slightly lower than the ENSANUT-ECU 2012.
It is necessary to remark that we observed a higher prevalence of markers that are present in the early stages of iron defi ciency anaemia, rather than the prevalence of anaemia itself, defi ned as low haemoglobin.One of these markers was RDW, which was described as a highly sensitive and useful tool for early diagnosis of mild iron defi ciency anaemia 21 .Also, it was recommended as a valuable tool for the determination of iron defi ciency anaemia in early stages.Furthermore, low MCV was even more frequent that high RDW in our population, something signifi cant because an isolated low MCV could indicate an initial phase of iron defi ciency by itself.However, these two fundamental tools, RDW and MCV, could frequently be missed during the paediatric evaluation 22 .The group of children under three years was most frequently aff ected by anaemia, a fi nding that could be explained because, through the fi rst 24 months of life, children have an increased risk to suff er anaemia, related to the active growth and the resulting increased need for iron per kilogram of body weight.Also, after six months of age, children need to get increasing quantities of iron from food, which is subject to various absorption limitations 23 .
Another notable fi nding was that approximately the half of patients included in our study were under three years of age, similarly as the globally observed.Additionally, there was a documented high mortality by respiratory tract infection in children less than fi ve years of age in developing countries, who are aff ected by poor nutritional status 24 .If an inadequate dietary intake is a risk factor to suff er pneumonia, then, iron defi ciency could also be suggested as a potential risk factor for respiratory tract infections.For example, a study conducted in Chile reported 20% of anaemia in children diagnosed with chronic respiratory diseases, just as we observed, approximately 20% of our patients diagnosed with pneumonia have anaemia.However, is necessary to remark this risk association in children under 32 months of age 25 .The signifi cant risk association between anaemia and pneumonia in children was also described in other studies.One of them conducted in the paediatric department of the Al-Azhar university hospital found a signifi cant decrease of RBCs, haemoglobin, ferritin, and iron in children diagnosed with pneumonia 26 .Additionally, a study which includes 200 infants and children between nine months to 16 years, found that anaemic subjects were 5.7 times more susceptible to low respiratory tract infections, as well as, what was described in a study conducted in the paediatric department of Manipal Teaching Hospital, which shows anaemic children were 3.2 times more susceptible to acute low respiratory tract infections (p<0.01) 27,28.A meta-analysis published in 2013, presents a series of risk factors for severe acute lower respiratory infections in children described by odds ratios (with 95% confi dence intervals): low birth weight 3.18 (1.02-9.90),lack of exclusive breastfeeding 2.34 (1.42-3.88),crowding-more than seven persons per household 1.96 (1.53-2.52),exposure to indoor air pollution 1.57 (1.06-2.31),incomplete immunization 1.83 (1.32-2.52),undernutrition-weight-for-age less than 2 standard deviations 4.47 (2.10-9.49),and HIV infection 4.15 (2.57-9.74).All, of these risk factors, should be considered in subsequent studies to clarify the importance of iron defi ciency anaemia as a risk factor for pneumonia 29 .In spite of these fi ndings, a study which includes 12 infants and children below fi ve years of age observed no infl uence of anaemia as a risk factor for pneumonia 30 .
The iron homeostasis is essential during immune responses as the reduction in its availability by proteins like hepcidin, decreases the bacterial growth.On the other hand, iron is required in cytokine production as well as participates in enzymatic phenomena like the activity of myeloperoxidase or even has been demonstrated quantitatively altered T cell subsets in iron defi ciency in children 31 .If we consider the importance of iron in immunology, is justifi ed in thinking that anaemia could be present in other infectious diseases.A study carried out in 293 children showed anaemia (Hb<11 g/dl) at six months was an independent risk factor for diarrhoea and respiratory disease in children with age between 7 to 18 months 32 .Similarly, in a previous study in which we studied the clinical characteristics of the urinary tract infection caused by Escherichia coli in the HE-1, in children between three months and 14 years of age, we found haemoglobin under 11.8 in 32.4% and 40%, of the patients admitted in the hospital and the external consultation, respectively 33 .
We found a similar time of hospital stay among diff erent conditions in our patients, something that could be explained because, with the antibiotic, pharmacological and no pharmacological support, the diff erence between the groups compared is reduced.This research has signifi cant limitations that should be considered; one of them is the reduced number of patients.So, it is necessary to expand the number of subjects.Another weakness is the point that we only considered data from our hospital for the analysis.Thus, it is essential to employ data from other health institutions for further studies to understand the infl uence of iron defi ciency in lower respiratory tract infection in our population.Also, as we did not include ferritin levels in each patient, it would be interesting to analyse all the wide range of iron defi ciency markers in children.

Conclusion
Our results suggest anaemia as a frequent condition in children diagnosed with pneumonia, particularly during the fi rst three years of life.Therefore, a detailed evaluation of patients in this group of age, which includes, nutritional assessment, is needed in both, hospital and primary health care.We suggest for new studies to study in detail the association between iron defi ciency markers and pneumonia, as well as other infectious diseases, especially in children under 36 months.It is essential also to consider the appropriate establishment of therapeutic algorithms to guarantee a rational use of antibiotics The anthropometric parameters were classifi ed as follows: weight for age (severe low weight [Z score<-3], low weight [Z score<-2], normal weight and high weight; height (severe low height [Z score<-3], low height [Z score<-2], normal and high height); and BMI (severely emaciated [Z score<-3], emaciated [Z score<-2], overweight [Z score>2] and obesity [Z score>3])

Table 1 :
Normal values of the haemogram used in this study

Table 2 :
Altitude adjustments according to place of residence Abbreviation; Masl: meters above sea level; HB: haemoglobin; g/l: grams per litre.

Table 3A :
Showing general characteristics of the groups and primary results of this work.

Table 3B :
Showing various anthropometric measurements

Table 4 :
Showing length of hospital stay Abbreviations; SD, standard deviation; max, maximum value; min, minimum value; HCT, haematocrit; MCV, mean corpuscular volume; MCH, mean corpuscular haemoglobin, RDW, red cell distribution width; The values of the haematological variables were assigned according to the reference range established in the methods section, *z-score obtained through Mann-Whitney U Test.