Prevalence and Correlates of Vitamin A Insufficiency Among 12–18-Month-Old Children Living in Slums of Mumbai, India

Objectives

We aimed to determine the prevalence of baseline vitamin A insufficiency (VAI; serum retinol (SR) < 1.05µmol/L) and to identify potential correlates of SR and VAI in a population of 12–18-month-old children participating in a randomized controlled trial in urban slums in Mumbai, India.

Methods

In this cross-sectional analysis, we determined SR concentrations in archived serum samples (N = 118) that were collected from 12–18-month-old children in urban slum communities of Western Mumbai in March–October 2017. We adjusted SR measurements using the BRINDA adjustment method for C-reactive protein.¹ Child, maternal and household characteristics were assessed as potential correlates. Linear [β(SE)] and binomial [RR (95% CI)] regressions were used to identify correlates of SR and VAI, respectively. Age and sex were retained in all models.

Results

The children in this population had a median (IQR) age of 14.5 (12.4, 16.7) months, and 46.6% were girls. Almost a third (27.1%) were underweight (weight-for-age Z-score <-2) and 9.3% were wasted (weight-for-length Z-score <-2). One third (30.5%) were anemic (hemoglobin < 11g/dL) and 19.5% of children were zinc-deficient (Zn < 70µg/dL). Median (IQR) unadjusted SR was 1.1 (0.9, 1.4) µmol/L and VAI was present among 50 (42.4%) of the population. After adjusting for inflammation, SR was 1.2 (1.0, 1.5) µmol/L and VAI was present in 38 (32.2%) of children. In multivariate regressions, each nmol/L increase in vitamin D [25(OH)D] was associated with a 0.01 µmol/L increase in SR [β (SE) 0.01 (0.004), p = 0.004], and each additional child under 5 years living in the household was associated with lower SR [−0.13 (0.04), p = 0.003].

Conclusions

In these 12–18-month-old children, VAI was prevalent in nearly two out of every five children.

Funding Sources

Division of Nutritional Sciences, Cornell University Harvest Plus.

Biomarkers of Choline Metabolism in Women of Reproductive Age in Southern India

Objectives

Inadequate maternal choline intake during pregnancy has been associated with adverse pregnancy and child health outcomes – and has been identified as a potential risk factor for neural tube defects. However, there is limited data on biomarkers of choline metabolism in women of reproductive age (WRA), and few representative population-level data from India. The objective of this analysis was to examine biomarkers of choline metabolism and their correlates in WRA, as part of a population-based biomarker survey in Southern India.

Methods

Participants were WRA (15–40y; n = 980) who were not pregnant or lactating. Free choline, betaine, dimethylglycine, methionine, and trimethylamine N-oxide concentrations were evaluated via liquid chromatography tandem mass spectrometry. Linear regression models were used to examine sociodemographic (e.g., age, parity), dietary (e.g., frequency of animal source food intake), and anthropometric (e.g., body mass index [BMI], waist circumference, waist-hip ratio) correlates of biomarkers of choline metabolism, including betaine concentrations and betaine to choline ratio (B:C).

Results

Betaine concentrations (GM: 44.2 μmol/L [95% CI: 43.4, 45.0]) were 4-fold higher compared to free choline (11.1μmol/L [10.9, 11.2]) levels, with an average B:C of 4.0 [3.9, 4.1]. Increased age (β: −0.01 [SE: 0.002], P < 0.0001), multiparity (multiparous/primiparous vs. nulliparous, P < 0.0001), and increased frequency of animal source food intake (poultry, p = 0.004; red meat, p = 0.01), were associated with lower B:C. Higher BMI (betaine: β: −0.004 [SE: 0.002], p = 0.02; B:C: β: −0.02 [SE: 0.002], P < 0.0001), waist circumference (betaine: −0.002 [0.001], p = 0.02; B:C: −0.01 [0.001], P < 0.0001), and waist-hip ratio (betaine: −0.31 [0.15], p = 0.04; B:C: −1.02 [0.16], P < 0.0001), were also associated with lower betaine concentrations and B:C.

Conclusions

In this population, higher BMI and central adiposity were associated with lower betaine concentrations and B:C. Findings from this biomarker survey will inform prospective research on the role of choline biomarkers in the health and nutritional requirements of WRA.

Funding Sources

Centers for Disease Control and Prevention (5U01DD001007), University of South Carolina Disability Research and Dissemination Center; Cornell DNS; NIH T32DK007158 (HMG); NIH 5T32HD087137 (AF).

Iron Status and Metabolic Health in Women of Reproductive Age in Southern India

Objectives

To examine the associations of iron status with metabolic health outcomes in women of reproductive age (WRA), as part of a population-based biomarker survey in Southern India.

Methods

Participants were WRA (15–40 y; n = 980) who were not pregnant or lactating (2017–2018). Blood samples were analyzed for hemoglobin (Hb; Coulter Counter) and glycated hemoglobin (HbA1c; nephelometry). Serum ferritin (SF) was measured by electrochemiluminescence; soluble transferrin receptor, C-reactive protein, and alpha-1 acid glycoprotein were analyzed via immuno-based assays (Roche). Anthropometric and blood pressure measurements were collected in triplicate. Bioelectrical impedance analysis (BC-418 MA) was used to estimate whole body (WF%) and trunk (TF%) fat in women ≥ 18y. Anemia was defined as Hb < 12.0 g/dL. SF was adjusted for inflammation using Biomarkers Reflecting Inflammation and Nutritional Determinants of Anemia (BRINDA) methods; iron deficiency (ID) was defined as SF < 15.0 µg/L. Linear and binomial regression were used to examine associations of iron status with metabolic outcomes.

Results

A total of 41.5% of WRA had anemia, and 46.3% had ID (61.5%; BRINDA-adjusted). A total of 23.3% of adults were overweight (BMI: 25.0-<30.0 kg/m²) and 9.6% had obesity (BMI ≥ 30.0 kg/m²). Waist circumference (WC; ≥88.9 cm) and waist-hip ratio (WHR; ≥0.85) were elevated in 13.4% and 20.8% of women, and 25.0% had elevated HbA1c (≥6.5%: 5.0%; ≥5.7-<6.5%: 20.0%). Higher Hb concentrations were associated with increased BMI (β: 0.42 [SE: 0.09]; p < 0.01), WC (0.77 [0.21]; p < 0.01), and WF% (0.89 [0.17]; p < 0.01). Higher SF levels were associated with higher WF% (β: 0.79 [SE: 0.32]; p = 0.01) and TF% (0.92 [0.39]; p = 0.02), and elevated WC (RR: 1.20 [95% CI: 1.02–1.42]). Iron status was not significantly associated with HbA1c or blood pressure.

Conclusions

The burden of adverse metabolic outcomes was substantial in this population. Higher iron status was associated with higher BMI and central adiposity. Evaluating iron status and metabolic outcomes in future studies could help inform screening and interventions to improve the health of WRA.

Funding Sources

Centers for Disease Control and Prevention (5U01DD001007), University of South Carolina Disability Research and Dissemination Center; Consejo Nacional de Ciencia y Tecnología (CEGL); NIH 5T32HD087137 (AF); NIH T32DK007158 (HMG).

Association Between Longitudinal Changes in Body Mass Index (BMI) and Body Fat With Cardiometabolic Risk Factors in Indian Children and Adolescents

Objectives

To evaluate the associations between longitudinal changes in body mass index (BMI) and body fat mass (FM%) with cardiometabolic risk factors in school-aged children and adolescents from South India.

Methods

This prospective school-based study comprised 4,289 children and adolescents (44.2% male) aged 5–17y participating in the PEACH-II and III studies in Bengaluru, India. Anthropometry and FM% (BOD POD, Cosmed; 5.2.0) were measured twice with a median age gap of 2.8y. Waist circumference (WC), systolic/diastolic blood pressure, HbA1c, and lipid profile were measured at endline. BMI-for-age z-scores were calculated using the World Health Organization Child Growth Standards, and were used to define overweight (≥+1 SD to < +2 SD) and obesity (≥+2 SD). Linear regression was used to examine the associations between longitudinal changes in BMI-for-age z-scores and FM% with cardiometabolic risk factors, after adjusting for parental BMI, socioeconomic status, age of menarche, inflammation (c-reactive protein), physical activity and night sleep duration.

Results

The prevalence of overweight/obesity and mean FM% increased from 15.2% to 20.5%, and 19.5% to 21.7%, respectively during follow-up. Overall, 43.8% and 15.5% of the participants had high triglycerides (TG, ≥200 mg/dL), and elevated HbA1C (≥5.7%), while 40.9% had low HDL (< 40 mg/dL); a total of 69.1% of participants had ≥ 1 abnormal biomarker(s). Increment in BMI for-age z-score of 1SD overtime was significantly associated with higher WC (β: 2.9, [SE:0.20] cm), TG (8.4, [1.3] mg/dL), HbA1c (0.17, [0.06] %), total cholesterol (1.6, [0.56] mg/dL), and lower HDL (−1.3, [0.26] mg/dL). Longitudinal increment in FM% was associated with higher WC (0.3, [0.03] cm), TG (1.2, [0.15] mg/dL) and total cholesterol (0.4, [0.1] mg/dL).

Conclusions

The burden of overweight/obesity and FM% substantially increased in this population of South Indian children and adolescents overtime, and were associated with cardiometabolic risk factors. Regular monitoring of BMI and body composition can help to identify children at risk of developing cardiovascular diseases. Appropriate dietary and lifestyle interventions for children at risk are needed in India, where chronic diseases and associated deaths are increasing at alarming rate in young and older adults.

Funding Sources

Funded by Indian Council of Medical Research.

Comparison of Anemia Screening Methods Using Paired Venous Samples in Women of Reproductive Age in Southern India

Objectives

To evaluate anemia screening via portable hemoglobinometer (HemoCue 301) compared to automated hematology analyzer (AHA) reference, using paired venous samples from a population-based biomarker survey in Southern India.

Methods

Participants were women of reproductive age (WRA; 15–40 y) who were not pregnant or lactating. Paired venous whole blood samples (n = 896) were analyzed for hemoglobin (Hb) via HemoCue 301 and Coulter Counter AHA, as the gold standard. Serum ferritin (SF) was measured by electrochemiluminescence. Anemia and severe anemia were defined as Hb &lt; 12.0 and &lt; 8.0 g/dL, respectively. SF was adjusted for inflammation using BRINDA methods; iron deficiency (ID) was defined as SF &lt; 15.0 μg/L. Bland-Altman methods were used to assess level of agreement between Hb results (i.e., mean difference, standard deviation of differences, limits of agreement). Diagnostic accuracy parameters (i.e., sensitivity, specificity, positive predictive value, negative predictive value, accuracy) were calculated to evaluate HemoCue performance compared to the AHA reference, overall and by sociodemographic (e.g., age, parity), nutritional (e.g., iron, vitamin B12, folate status), and metabolic (e.g., HbA1c, body mass index (BMI)) characteristics.

Results

The estimated prevalence of anemia (36.3% vs. 41.6%, p &lt; 0.0001) was significantly lower via HemoCue compared to the AHA reference. A total of 84.4% of anemia and 98.8% of severe anemia screening results were concordant. HemoCue had 74.8% sensitivity and 91.2% specificity, compared to AHA. HemoCue sensitivity was higher in women with iron deficiency (SF &lt; 15.0 μg/L: 81.6% vs. ≥15.0 μg/L: 41.3%; p &lt; 0.0001), and lower in WRA who were overweight (BMI ≥ 25.0 kg/m2: 63.9% vs. &lt; 25.0 kg/m2: 78.8%; p = 0.004) or had elevated C-reactive protein (CRP) (CRP &gt; 1.0 mg/L: 67.2% vs. ≤1.0 mg/L: 82.9%; p = 0.001; CRP &gt; 3.0 mg/L: 62.2% vs. ≤3.0 mg/L: 79.3%; p = 0.001).

Conclusions

Findings suggest the estimated prevalence of anemia evaluated via portable hemoglobinometer (HemoCue) was lower compared to the AHA reference. HemoCue sensitivity varied by iron status and metabolic risk factors in this population.

Funding Sources

Centers for Disease Control and Prevention (5U01DD001007), University of South Carolina Disability Research and Dissemination Center; NIH 5T32HD087137 (AF); NIH T32DK007158 (HMG); Cornell DNS.

A Randomized Trial of Iron-Biofortified Pearl Millet-Based Complementary Feeding in Children Aged 12 to 18 Months Living in Urban Slums in India

Objectives

To determine the efficacy of iron-biofortified pearl millet (FePM, Dhanashakti, ICTP-8203Fe)-based complementary feeding on iron status and growth, in children living in urban slums of Mumbai, India.

Methods

This randomized controlled feeding trial of FePM was conducted among 223 children aged 12 to 18 months who were not severely anemic (hemoglobin (Hb) ≥9.0 g/dL) (Clinicaltrials.gov ID: NCT02233764). Children were randomized to receive either complementary foods prepared from FePM or conventional non-biofortified pearl millet (CPM) daily for 9 months. Iron status (Hb, serum ferritin (SF)), and anthropometric indicators (i.e., length, weight, head, mid-upper arm circumferences, triceps and subscapular skinfolds) were evaluated at enrollment and throughout the trial. World Health Organization (WHO) anthropometric Z-scores were calculated using WHO growth standards. Primary outcomes were Hb and SF concentrations, and growth, defined as WHO Z-scores. An intent to treat approach was used for analyses. We used the Hodges–Lehmann–Sen test to assess the change in primary outcomes between baseline and the last visit and report corresponding 95% confidence intervals.

Results

At baseline, 67.7% of children were anemic (Hb < 11.0 g/dL) and 59.6% were iron-deficient (SF < 12.0 µg/L). FePM did not significantly increase iron biomarkers or improve growth, compared to CPM. In subgroup analyses, FePM improved hemoglobin concentrations in male children, and in children with iron deficiency or iron depletion (SF < 25.0 µg/L) at baseline, compared to CPM.

Conclusions

Daily consumption of FePM-based complementary foods did not significantly improve iron status or growth in children living in this population. However, the intervention significantly improved hemoglobin concentrations among male children and individuals who were iron-deficient or iron-depleted at baseline.

Funding Sources

HarvestPlus.

Longitudinal Changes in Waist Circumference and Waist-to-Height Ratio in Children and Adolescents in Southern India

Objectives

Waist-to-height ratio has been proposed as a surrogate indicator to screen for abdominal obesity in children. The objective of this analysis was to examine longitudinal changes in waist circumference and waist-to-height ratio in school-aged children and adolescents in Southern India.

Methods

Participants were 3,955 children and adolescents (6–16 y) participating in a prospective school-based study in Bangalore, India (PEACH-II and III). Anthropometry, including weight, height, and waist circumference (WC), were collected at baseline and endline, with a median follow up of 2.9 years [IQR: 2.0–4.0 y]. Elevated waist circumference was defined as ≥75^th percentile, based on age- and sex-specific percentile curves developed for this population. Waist-to-height ratio (WHtR) was calculated as waist circumference (cm) divided by height (cm), and elevated WHtR was defined as ≥0.5. Body mass index (BMI)-for-age z-scores were calculated using the World Health Organization (WHO) Child Growth Standards, to define overweight (≥+1 to < +2 SD) and obesity (≥+2 SD). Paired t-tests and McNemar test were used to examine changes in WC and WHtR during follow-up.

Results

At baseline, a total of 11.4% participants were overweight and 4.1% were obese. The prevalence of elevated WC increased from 15.4% to 27.1% during follow-up (P < 0.001), with an average increase of 9.92 [SD: 8.46] cm. The prevalence of elevated WC increased from 9.2% to 26.3% (P < 0.001) among children who were normal weight at baseline, and 60.0% to 75.3% (P < 0.001) among children who were overweight. The prevalence of elevated WHtR increased from 9.6% to 20.5% during follow-up (P < 0.001), with the greatest increases in children who were normal weight (2.9% to 17.6% [P < 0.001]) or overweight (42.0% to 62.5% [P < 0.001]) at baseline.

Conclusions

The prevalence of elevated WC and WHtR was substantial in this population and increased over time. Findings suggest that WHtR, which is easy to measure and interpret, could be used in addition to WC during routine school-based examination to screen for risk of overweight and central adiposity.

Funding Sources

Partially funded by Indian Council of Medical Research (ICMR); Fulbright Commission and National Research and Development Agency (ANID) of Chile fellowship (XEPM).

Vitamin B12 Status and Metabolic Health in Women of Reproductive Age in Southern India

Objectives

To examine the burden of metabolic outcomes and associations of vitamin B12 status with metabolic health in women of reproductive age (WRA), as part of a population-based biomarker survey in Chittoor, India.

Methods

Participants (980 WRA; 15–40y nonpregnant or lactating) were assessed for glycated hemoglobin (HbA1c; nephelometry) and serum vitamin B12 concentrations (chemiluminescence). Anthropometric measurements and systolic (SBP) and diastolic (DBP) blood pressures were collected in triplicate. Bioelectrical impedance analysis was used to evaluate whole body (WF%) and trunk (TF%) fat among women ≤ 18y. We defined elevated HbA1c as ≤ 6.5% and ≤ 5.7-&lt; 6.5%, and hypertension as stage 1 (SBP 130–139 or DBP 80–89 mmHg) and stage 2 (SBP ≤ 140 or DBP ≤ 90 mmHg). Vitamin B12 was natural logarithmically transformed prior to analyses; vitamin B12 deficiency

was defined as &lt; 148 pmol/L. Linear and binomial regression models were used to examine associations of vitamin B12 status with metabolic outcomes.

Results

A total of 23.3% of adult WRA were overweight (body mass index (BMI): 25.0 to &lt; 30.0 kg/m2) and 9.7% had obesity (≤30.0 kg/m2). Waist circumference (WC; ≤88.9 cm) and waist-hip ratio (WHR; ≤0.85) were elevated in 13.4% and 20.1% of adult WRA. One-fourth of WRA had elevated HbA1c (≤6.5%: 5.0%; ≤5.7-&lt; 6.5%: 20.0%), and 18.6% had hypertension (stage 1: 16.4%; stage 2: 2.2%); 48.3% of WRA were vitamin B12 deficient. Higher continuous vitamin B12 concentrations were associated with lower BMI (β [standard error (SE)] -0.65 [0.28]) and WF% (-1.01 [0.50]); lower risk of elevated WC (risk ratio (RR) [95% confidence interval] 0.64 [0.49–0.85]); and higher risk of HbA1c ≤ 5.7% (1.19 [1.00–1.41]). Vitamin B12 deficiency was associated with higher BMI (β [SE] 0.98 [0.34], p = 0.004), WC (1.96 [0.76]), WF% (1.75 [0.59]), and TF% (2.03 [0.73]); and higher risk of having overweight (RR: 1.31 [1.09–1.58]), elevated WC (1.85 [1.32–2.60]), and WHR (1.38 [1.07–1.78]).

Conclusions

The burden of adverse metabolic outcomes was substantial in this population, and vitamin B12 deficiency was associated with central adiposity and overweight. Evaluating the role of vitamin B12 in the development of metabolic outcomes in future studies could inform screening and interventions to improve vitamin B12 status and metabolic health in WRA.

Funding Sources

Centers for Disease Control and Prevention.

Micronutrient Status and Child Growth in Mexico: Results From the 2018 National Health and Nutrition Survey (ENSANUT)

Objectives

To examine the associations between micronutrient status and child growth, as part of the National Health and Nutrition Survey in Mexico.

Methods

Data from the National Health and Nutrition Survey (ENSANUT 2018) were analyzed to examine the association between micronutrient status and child growth in 1,656 children 36 to 59 months of age. Micronutrient biomarkers (serum ferritin (SF), vitamin B12, vitamin D, serum zinc) were measured in venous blood samples via immunoassays. Child weight and height were measured using standardized procedures, and indicators weight-for-age (WAZ), height-for-age (HAZ), weight-for-height (WHZ), and body mass index (BMI) Z-scores were calculated using the WHO Anthro V3.1 and categorized as per WHO Child Growth Standards. Micronutrient biomarkers were natural logarithmically transformed prior to analyses. Iron deficiency was defined as SF &lt; 12.0 μg/L and iron insufficiency was defined as SF &lt; 20.0 μg/L. Vitamin B12 deficiency and insufficiency were defined as &lt; 148 and &lt; 221 pmol/L, respectively. Vitamin D deficiency was defined as &lt; 50 nmol/L, and zinc deficiency was defined as &lt; 65μg/dL. Survey linear regression models were used to examine associations of micronutrient status with WHO Z-scores.

Results

Serum ferritin concentrations (β [SE]: 0.096 [0.001], p &lt; 0.0001) and vitamin B12 (0.092 [0.003], p &lt; 0.0001) concentrations were associated with higher WHZ. Vitamin B12 (0.171 [0.004], p &lt; 0.0001), vitamin D (0.358 [0.007], p &lt; 0.0001), and zinc (0.122 [0.006], p &lt; 0.0001) concentrations were associated with higher HAZ. Iron deficiency (β [SE]: -0.469 [0.004], p &lt; 0.0001) and zinc deficiency (-0.203 [0.01], p &lt; 0.0001) were associated with lower WHZ, and vitamin B12 (-0.449 [0.002], p &lt; 0.0001), vitamin D (-0.215 [0.001], p &lt; 0.0001) and zinc (-0.216 [0.01], p &lt; 0.0001) deficiencies were associated with lower HAZ.

Conclusions

Micronutrient deficiencies of vitamin B12, vitamin D, and zinc were associated with lower linear growth in young children in Mexico.

Funding Sources

ENSANUT was funded by The Ministry of Health of Mexico.

Estimating the Serum Folate Concentration Associated With the Red Blood Cell Folate Threshold for Optimal Neural Tube Defects Prevention: A Population Based Biomarker Survey

Objectives

To estimate the serum folate insufficiency threshold (sf-IT) corresponding to the red blood cell (RBC) folate insufficiency threshold for optimal neural tube defect (NTD) prevention.

Methods

Participants were 977 women of reproductive age (WRA; 15–40y; not pregnant or lactating) from a population-based biomarker survey in Southern India. Venous blood samples were collected at enrollment. Plasma, serum, and red blood cells were centrifuged, processed, and stored &lt; -80°C until batch analysis. Total vitamin B12 concentrations were measured via chemiluminescence; RBC and serum folate concentrations were measured using the World Health Organization-recommended microbiological assay. Vitamin B12 deficiency was defined as total vitamin B12 &lt; 148 pmol/L. Folate

insufficiency was defined as RBC folate &lt; 748 nmol/L, the recommended calibrator-adjusted equivalent of the threshold for population optimal NTD prevention. A previously developed Bayesian model and the RBC and serum folate distributions in this population were used to estimate the sf-IT corresponding to the RBC folate insufficiency threshold for optimal NTD prevention, overall and by age, body mass index (BMI) category, HbA1c, anemia, and vitamin B12 status.

Results

The overall estimated median sf-IT was 37.8 nmol/L (95% credible interval [33.8–43.3]). This threshold was lower in overweight WRA (BMI: ≥25.0 kg/m2: 32.0 nmol/L [27.3–40.2] vs. BMI &lt; 25.0 kg/m2: 36.2 nmol/L [32.2–43.3]), and varied by age (&lt; 25y: 61.3 nmol/L [44.3–111.8]; 25 to 35y: 35.7 nmol/L [30.8–43.5]; ≥35y: 30.8 nmol/L [26.9–37.2]). The sf-IT was lower in anemic WRA (32.9 nmol/L [28.5–40.1]) compared to non-anemic WRA (42.0 nmol/L [36.1–51.3]), and lower in WRA with elevated HbA1c (≥5.7% to &lt; 6.5: 32.4 nmol/L [27.3–41.6]; ≥6.5%: 20.9 nmol/L [17.8–25.6]) vs. WRA with HbA1c &lt; 5.7% (43.8 nmol/L, [37.5–53.7]). The median sf-IT was higher in WRA with vitamin B12 deficiency (72.1 nmol/L [52.0–126.0]), compared to women who were not vitamin B12 deficient (28.1 nmol/L [25.6–31.5]).

Conclusions

The estimated sf-IT is dependent on anemia, elevated HbA1c, BMI, age, and vitamin B12 status.

Funding Sources

Centers for Disease Control and Prevention; AF was supported by the National Institutes of Health #5 T32 HD087137.

Micronutrient Deficiencies and Anemia in Children in Mexico: Results From the National Health and Nutrition Survey (ENSANUT)

Objectives

To examine the burden of anemia and micronutrient deficiencies, and associations between micronutrient status and anemia in children (1–11 y) in Mexico.

Methods

Data from the National Health and Nutrition Survey (ENSANUT 2018) were analyzed to determine the burden of anemia and micronutrient deficiencies (iron, vitamin B12, vitamin A) in a nationally-representative sample of 4,891 children 1 to 11 years of age (1 to &lt; 5 y: 1,337; 5–11 y: 3,554). Hemoglobin (Hb) was measured via HemoCue and venous blood samples were collected, processed, and stored at &lt; -80°C until analysis. Serum ferritin (SF), C-reactive protein (CRP) and vitamin B12 concentrations were measured via immunoassays. Serum retinol was measured via HPLC. Hemoglobin was adjusted for altitude; anemia (&lt; 5 y: Hb &lt; 11.0 g/dL; 5–11 y: &lt; 11.5 g/dL) and iron deficiency (ID; &lt; 5 y: &lt; 12.0 μg/L; 5–11 y: &lt; 15.0 μg/L) were defined using age-specific cut-offs. Iron

insufficiency was defined as SF &lt; 20.0 μg/L, and inflammation as CRP &gt; 5.0 mg/L. Vitamin B12 deficiency and insufficiency were defined as &lt; 148 and &lt; 221 pmol/L, and vitamin A deficiency was defined as retinol &lt; 20 μg/dL. Survey linear and logistic regression were used to examine associations of micronutrient biomarkers with hemoglobin concentrations and odds of anemia.

Results

Serum ferritin concentrations (β [SE]: 0.0003, p &lt; 0.001) and serum retinol levels (0.0005, p &lt; 0.001) were associated with higher hemoglobin levels, while iron deficiency (–0.57 [0.03], p &lt; 0.001), iron insufficiency (–0.54 [0.02]; p &lt; 0.001), and vitamin A deficiency (–0.53 [0.01], p &lt; 0.0001) were associated with lower hemoglobin levels. In analyses of anemia, the odds of anemia were 11% lower per 10 μg/L increase in serum ferritin (OR [95% CI]: 0.89 [0.84–0.93], p &lt; 0.001) and 34% lower per 10 μg/dL increase in serum retinol (OR: 0.66 [0.57–0.77], p &lt; 0.001), while iron deficiency (OR: 2.22 [1.49–3.29], p &lt; 0.001) and iron insufficiency (OR: 1.74 [1.32–2.30]; p = 0.0001) were associated with higher odds of anemia.

Conclusions

Findings suggest that the burden of anemia and micronutrient deficiencies are high in children in Mexico, and lower iron and vitamin A status were associated with increased odds of anemia.

Funding Sources

ENSANUT was funded by The Ministry of Health of Mexico.

Virtual reality as a learning tool in spinal anatomy and surgical techniques

Background

Surgical simulation is a valuable educational tool for trainees to practice in a safe, standardized, and controlled environment. Interactive feedback-based virtual reality (VR) has recently moved to the forefront of spine surgery training, with most commercial products focusing on instrumentation. There is a paucity of learning tools directed at decompression principles. The purpose of this study was to evaluate the efficacy of VR simulation and its educational role in learning spinal anatomy and decompressive techniques.

Methods

A VR simulation module was created with custom-developed software. Orthopaedic and neurosurgical trainees were prospectively enrolled and interacted with patient-specific 3D models of lumbar spinal stenosis while wearing a headset. A surgical toolkit allowed users to perform surgical decompression, specifically removing soft tissues and bone. The module allowed users to perform various techniques in posterior decompressions and comprehend anatomic areas of stenosis. Pre- and post-module testing, and utility questionnaires were administered to provide both quantitative and qualitative evaluation of the module as a learning device.

Results

28 trainees were enrolled (20-orthopaedic, 8-neurosurgery) in the study. Pre-test scores on anatomic knowledge progressively improved and showed strong positive correlation with year-in-training (Pearson's r = 0.79). Following simulation, the average improvement in post-test scores was 11.4% in junior trainees (PGYI-III), and 1.0% in senior trainees (PGYIII-Fellows). Knowledge improvement approached statistical significance amongst junior trainees (p = 0.0542). 89% of participants found the VR module useful in understanding and learning the pathology of spinal stenosis. 71% found it useful in comprehending decompressive techniques. 96% believed it had utility in preoperative planning with patient-specific models.

Conclusions

Our original VR spinal decompression simulation has shown to be overwhelmingly positively received amongst trainees as both a learning module of patho-anatomy and patient-specific preoperative planning, with particular benefit for junior trainees.

Anemia and Iron and Vitamin B12 Deficiencies in Children Under 5 in Mexico: Results from the National Health and Nutrition Survey (ENSANUT)

Objectives

To examine the burden of anemia, iron deficiency, and vitamin B12 deficiency in children under 5 in Mexico.

Methods

Data from the recently completed National Health and Nutrition Survey (ENSANUT 2018) were analyzed to examine the burden of anemia and micronutrient deficiencies in children under 5 (1 to &lt;5 years). Hemoglobin (Hb) concentrations werequantified in capillary blood samples (n = 3144) via HemoCue.Venous blood samples (n = 1019) were collected, centrifuged, processed, and stored &lt;−80°C until analysis of micronutrient biomarkers. Serum ferritin (SF), C-reactive protein (CRP), and vitamin B12 concentrations were measured via immunoassays. Hemoglobin was adjusted for altitude; anemia was defined as hemoglobin &lt;11.0 g/dL. Iron deficiency was defined as serum ferritin &lt;12.0 µg/L and iron insufficiency was defined as serum ferritin &lt;20.0 µg/L; inflammation was defined as CRP concentrations &gt; 5.0 mg/L (and CRP &gt; 1.0 mg/L). Iron deficiency anemia was defined as hemoglobin &lt;11.0 g/dL and serum ferritin &lt;12.0 μg/L. Vitamin B12 deficiency and vitamin B12 insufficiency were defined as &lt;148.0 and &lt;221.0 pmol/L, respectively. Survey frequency procedures were used to examine the prevalence of anemia and micronutrient deficiencies; survey linear and logistic regression were used to examine associations of micronutrient biomarkers with hemoglobin concentrations and odds of anemia.

Results

A total of 22.9% of children were anemic in this population. In the biomarker sub-sample, 10.0% of children were iron deficient (SF &lt; 12.0 µg/L), and 29.9% had serum ferritin concentrations &lt;20.0 µg/L. A total of 9.5% of children had CRP concentrations &gt;5.0 mg/L, and 34.7% had CRP &gt; 1.0 mg/L. The prevalence of vitamin B12 deficiency (vitamin B12 &lt; 148.0 pmol/L) was 5.4% and 21.5% of children had vitamin B12 insufficiency (vitamin B12 &lt; 221.0 pmol/L).

Conclusions

Findings suggest that the burden of anemia and iron and vitamin B12 insufficiency is high in young children in Mexico.

Funding Sources

ENSANUT was funded by The Ministry of Health of Mexico.

Nutrient-Dense Meal Delivery in Partnership with Small-Scale Producers in Mumbai Urban Slums: Implementation Considerations Within a Randomized Controlled Feeding Trial

Objectives

Despite the multitude of evaluations of nutrition intervention trials in low-income settings, documented practice to operate such trials is scarce. We aimed to fill this gap by outlining key steps and resources required to run a small-scale intensive feeding trial such as establishing collaborations, developing an operational infrastructure, and daily orchestration of logistics for food preparation, delivery, and administration.

Methods

A randomized controlled feeding trial among 223 children in urban slums of Mumbai, India was completed in 2018. We established partnerships to facilitate the acquisition and proper storage of the study intervention, biofortified and conventional pearl millet. Feeding centers (FC) were identified based on a census. Two staff per FC managed daily operations: weighing and dispensing individual portions to participants based on their group allocation, recording of leftovers and adverse events.

The study team developed and standardized recipes for all menu items. Development considered staple crop and nutrient density per serving, cost and availability of ingredients, and scalability. Highly accepted recipes were included in a cyclic menu. Hot meals were prepared in partnership with the university canteen, where a kitchen coordinator was responsible for daily quality control, aliquoting, and dispensation to contracted auto-drivers for FC delivery. A certified bakery produced customized shelf stable items in bulk. Items were vacuum packed individually, labeled, and delivered to FC directly. Shelf-life was tested periodically.

Results

Over 26 tons of crops were procured at a cost of 25,000 USD. The cost of preparing and delivering about 30,000 fresh meals was about 15,000 USD, while the cost of approximately 60,000 shelf stable meals was about 10,000 USD. Over 15 months, we served 91,815 meals at a cost of 0.59 USD/meal in the context of a randomized controlled feeding trial.

Conclusions

We highlight opportunities and challenges of developing a sustainable food supply system for an intervention trial. We also outline a sustainable model for delivery of nutrient-dense meals in partnership with small-scale producers instead of establishing a centralized kitchen to facilitate advocacy and scale up.

Funding Sources

HarvestPlus.

Presence of Ebola Virus in Breast Milk and Its Risk of Transmission to Breastfeeding Infants: Synthesis of Evidence

Objectives

This study aimed to synthesize evidence related to the presence in and transmission of the Ebola virus (EBOV) through breast milk to help inform global guidelines on infant feeding.

Methods

We conducted a comprehensive systematic search in international and regional databases to identify original studies describing women with suspected or confirmed EBOV infection intending to breastfeed or give breast milk to an infant. Studies documenting any breastfeeding woman who has been vaccinated or is suspected or confirmed of viral infection at any time during or after pregnancy were identified. Records were independently screened by two authors and after duplicate records were removed, studies that met the inclusion criteria were selected for data extraction.

Results

From a total of 24,473 non-duplicate studies, we found six case reports that included seven breastfeeding mothers with suspected or confirmed EBOV infection and their eight children (one mother with twins). EBOV was detected via RT-PCR and/or by culture in five out of six breast milk samples that were collected from the mothers. Five out of the seven breastfed infants were found positive for EBOV infection and all of the identified cases led to death.

Conclusions

Ebola virus was detected in 5 out of 6 breast milk samples analyzed. Since the virus has also been detected in tears, saliva, and sweat, it is not possible to conclude with certainty that the transmission was through breast milk. Prospective studies are needed in order to define the safety of feeding infants with breast milk from mothers infected with EBOV.

Funding Sources

Department of Nutrition and Food Safety, WHO.

Biomarkers of Inflammation in the Context of Acute Illness Among Children 12–24 Months of Age Living in the Urban Slums of Mumbai

Objectives

To examine and compare the concentration of acute phase proteins, including nutritional and inflammation biomarkers, by morbidity status among children 12–24 months old living in urban slums of Mumbai.

Methods

12–24-month-old children were screened for participation in a randomized controlled trial involving feeding of iron- and zinc-biofortified pearl millet complementary foods. Information on acute illness including fever, respiratory symptoms, and diarrhea on the day of screening, in the past 7 days, and the last 4 weeks was collected along with blood samples for analysis of concentrations of nutritional (ferritin, retinol) and inflammation biomarkers (C-reactive protein, alpha-1-glycoprotein) involved in the acute phase response. Difference in medians were compared by the Kruskal-Wallis Test.

Results

The mean age of children was 14.9 months; 51% were males. More than half the children (54%) had at least one or more symptoms such as fever, cough, or diarrhea at any timepoint in the 4 weeks prior to the interview. Among those with symptoms, fever and respiratory symptoms, either alone or in combination, were most common (44.8%), while diarrhea (alone or in combination with fever or respiratory symptoms) was present in only 11% of the population. More than a quarter had some illness in the 7 days before the interview (25.8%) with 2.5% having illness (fever and/or respiratory symptoms) on the day of the interview. Among those reporting no morbidity, 7% had elevated CRP. Among those reporting acute illness on the day of interview or in the past 7 days, 13% had elevated CRP. When compared across different kinds of acute illness and having no symptoms, there was no significant difference in the median concentrations of CRP, ferritin, and retinol. Similar results were observed on comparing the concentration of these acute phase reactants across children with no, one, two, or more symptoms.

Conclusions

Among this pediatric population, we did not observe any significant difference in the concentration of acute phase reactants by illness type or number of concurrent illnesses compared with apparently healthy children.

Funding Sources

HarvestPlus.

Prevalence of Undernutrition and Anemia in 6–24-Month-Old Children in Rural South India: Assessing the Target Population for a Multiple Biofortified Crops Feeding Trial

Objectives

To characterize anemia and anthropometry in children in rural South India:1.Report the prevalence of undernutrition and anemia2.Examine associations with demographic, socio-economic, and clinical characteristics

among 6–24-month-old children screened for participation in a randomized controlled feeding trial.

Methods

This cross-sectional assessment included 339 children screened for participation in a randomized controlled feeding trial from March – September 2019. Complete blood counts were performed immediately after blood and anthropometric measurements were collected. Sociodemographic data were collected from mothers. We assessed the outcomes hemoglobin (Hb, g/dL), anemia (Hb &lt; 11 g/dL), stunting (LAZ &lt; −2), wasting (WLZ &lt; −2) and underweight (WAZ &lt; −2) with correlates at the individual (child age, sex, birthweight), maternal (maternal age, maternal education) and household (family income) level. Covariates were included based on previous literature and biological plausibility. Linear regression (ß, SE) was used to determine covariates for continuous outcomes (such as Hb) and binomial and Poisson regressions (RR, 95% CI) were used in the case of categorical outcomes (Hb &lt; 11 g/dL, LAZ &lt; −2, WLZ &lt; −2, WAZ &lt; −2). All multivariate models included a priori determined covariates age and sex and covariates with univariate p-values ≤ 0.20. Variables with a p-value of ≤ 0.05 were retained in the final multivariate models (SAS 9.4).

Results

Almost half (46.9%) of this population was ≤ 12 months of age at the time of assessment. Over half (52.7%) of participants were female. Anemia was present in 72.6% of children. 15.1% of children were stunted, 14.2% underweight and 8.2% were wasted based on the corresponding WHO Z-scores. A higher birth weight of 1 kg was associated with a lower risk of later underweight [0.47 (0.29, 0.79); P = 0.01] and wasting [0.42 (0.21, 0.85); P = 0.02]. Children with families whose monthly income was ≤ 5000 INR (∼USD 70) per month were more likely to be stunted [2.32 (1.17, 4.59); P = 0.01] and underweight [2.07 (1.05, 4.06); P = 0.03].

Conclusions

In these 6–24-month-old children, anemia and poor growth are widely prevalent. The results from this analysis provide evidence of a potential to benefit from nutritional interventions targeted at children in this population.

Funding Sources

Cornell Division of Nutritional Sciences Harvest Plus.

Nutritional Status and Measles Antibody Titer in Children Living in Urban Slums of Mumbai

Objectives

To determine a) the prevalence of protective measles IgG titers in young children living in urban slums of Mumbai, and b) the association of nutritional status assessed through a combination of anthropometry and micronutrient biomarker concentrations with these titers at baseline and after 9 months of follow-up.

Methods

Anthropometry and serum concentrations of vitamins A, B12, D; ferritin, and zinc, and anti-measles immunoglobulin G (IgG) were measured among 12–24-month-old children in urban slums of Mumbai. Demographic data and health history were also collected. The association between baseline predictors and measles titers was assessed using linear regression for continuous measles titer and change in measles titer, and binomial regression for the dichotomous outcome of having a protective titer at baseline and at endpoint (&gt;16.5 AU/mL).

Results

At baseline (n = 122, median age 14.5 months), 71% of children reported receiving dose 1 of the measles vaccine typically administered at 9 months of age, while 57% of children had protective measles IgG titers (&gt;16.5 AU/mL). After 9 months of follow-up, 73% of children had protective titers. Multivariate models found measles IgG titer at baseline was inversely associated with vitamin D deficiency (&lt;50 nmol/L) and vitamin B12 deficiency (&lt; 148 pmol/L), and positively associated with child's age, maternal height, and birth weight. Iron deficiency (serum ferritin &lt;15 ng/ml) at baseline was associated with lower measles IgG titers at endpoint (median age, 23.9 months).

Conclusions

The prevalence of protective antibody titer against measles virus in this pediatric urban slum population was less than anticipated, and age, maternal height, and birth weight were positively associated with increasing measles IgG titers. Deficiency in vitamins A, B12, and D, as well as iron, were associated with lower measles IgG titers, highlighting the potential role of micronutrient status in vaccine response.

Funding Sources

HarvestPlus.

Anemia and Micronutrient Deficiencies in Young Children in Urban Slums of Mumbai, India

Objectives

To examine the burden of anemia and micronutrient deficiencies among young children living in urban slums of Mumbai, India.

Methods

Participants were children 10–18 months of age (n = 278) residing in urban slums in Mumbai who were screened as part of a randomized feeding trial of biofortified pearl millet. Venous blood was collected by a pediatric phlebotomist and hemoglobin (Hb) concentrations were assessed via Coulter counter. Samples were processed and stored &lt;−80°C until laboratory analyses. Serum ferritin (SF), vitamin B12, and erythrocyte (RBC) folate concentrations were analyzed via chemiluminescence. C-reactive protein (CRP) was analyzed by nephelometry, α-1 acid glycoprotein (AGP) was analyzed by spectrophotometry, and serum retinol was evaluated via HP-LC. Serum ferritin and serum retinol levels were adjusted for inflammation using methods proposed by Thurman. Anemia and severe anemia were defined as Hb &lt; 11.0 g/dL and Hb &lt; 7.0 g/dL. Iron deficiency (ID) was defined as SF &lt; 12.0 µg/L, iron insufficiency was defined as SF &lt; 20.0 µg/L, and inflammation was defined as CRP &gt; 5.0 mg/L. Iron deficiency anemia was defined as Hb &lt; 11.0 g/dL and SF &lt; 12.0 µg/L. Vitamin A deficiency was defined as serum retinol &lt;20.0 µg/L. Vitamin B12 deficiency and vitamin B12 insufficiency were defined as &lt;148.0 pmol/L and &lt;221.0 pmol/L; folate deficiency was defined as RBC folate &lt;340 nmol/L. Linear and binomial regression models were used to examine associations of micronutrient biomarkers with Hb levels and risk of anemia.

Results

A total of 75.5% of children were anemic and 2.9% had severe anemia. The prevalence of iron deficiency was 61.4% (69.5% after Thurman adjustment), 57.4% of children had IDA, and 9.7% had CRP &gt; 5.0 mg/L. The prevalence of vitamin B12 deficiency was 17.0%, 41.8% of children had vitamin B12 insufficiency, and 19.9% had folate deficiency. Among anemic children, 84.5% had ≥1 micronutrient deficiency and 22.7% had ≥ 2 micronutrient deficiencies. Iron deficiency (RR: 2.15; 95% CI: 1.55, 2.99; P &lt; 0.001), and vitamin B12 deficiency (RR: 1.32; 95% CI: 1.09, 1.60; P &lt; 0.01) were associated with significantly increased risk of anemia.

Conclusions

Findings indicate that the burden of anemia, iron deficiency, and vitamin B12 insufficiency were high in young children in Mumbai, India.

Funding Sources

Harvest Plus; AF was supported by the National Institutes of Health.

Maternal reproductive history and premenopausal risk of hypertension and cardiovascular disease: a Danish cohort study

OBJECTIVES

The metabolic changes of pregnancy resemble a cardiovascular risk profile and may persist postpartum, with body mass index (BMI) as a potential modifier. We examined the association between the number of live-birth pregnancies and maternal premenopausal risks of hypertension and cardiovascular disease (CVD), accounting for pre-pregnancy BMI as well as abortions and stillbirths.

DESIGN

A prospective cohort study.

SETTING

Mothers from the Danish National Birth Cohort (1996 to 2002). For each of these women, registry data on all pregnancies from 1973 to 2011 were obtained, as were self-reported pre-pregnancy weight and height.

PARTICIPANTS

A total of 44 552 first-time mothers in the Danish National Birth Cohort.

OUTCOME MEASURES

Risks of hypertension and CVD during and between live-birth pregnancies separately and combined as live-birth cycles.

RESULTS

After adjustment for abortions, stillbirths, pre-pregnancy BMI and other covariates, a higher risk of hypertension was observed in the first (HR 1.53, 95% CI: 1.37 to 1.72) and fourth and subsequent live-birth cycles (HR 1.72, 95% CI: 1.15 to 2.58), compared with the second. However, as number of live-birth pregnancies increased, risk of hypertension decreased during live-birth pregnancies and increased between live-birth pregnancies (tests for trend, p<0.01). For CVD, we found an overall J-shaped but non-significant association with number of live-birth pregnancies. No interaction with pre-pregnancy BMI (<25 versus ≥25 kg/m²) was observed.

CONCLUSIONS

Premenopausal women had the highest risk of hypertension and CVD during their first live-birth pregnancy and after their fourth live-birth pregnancy. All risks were independent of BMI before the first live-birth pregnancy and of number of abortions and stillbirths.

Malnutrition and Suspected Dengue Virus Infection in Children in Coastal Ecuador (P10-120-19)

Objectives

To examine the burden of anemia and malnutrition in children presenting with suspected dengue virus infection, as part of our ongoing surveillance program for acute febrile illnesses in Ecuador.

Methods

A total of 58 children (6 months to 5 y) were recruited in 2018 as part of the surveillance program. Sociodemographic characteristics, clinical signs and symptoms, and anthropometric data were obtained using structured interviews and standardized assessments. Venous blood samples were collected, and hemoglobin was assessed via the sodium lauryl sulfate method. Suspected dengue cases were identified using the World Health Organization dengue severity classification guidelines. Anemia was defined as hemoglobin <110 g/L. Stunting, underweight, and wasting were defined based on their corresponding z-scores: height-for-age z-score (HAZ) <-2, weight-for-age z-score (WAZ) <-2, and weight-for-length z-score (WHZ) <-2, respectively. Overweight and obesity were defined as BMIZ >1 SDs, and BMIZ >2 SDs, respectively. Binomial and linear regressions were used to examine the associations of hemoglobin, anemia, and anthropometric variables with clinical variables among patients presenting with suspected dengue infection.

Results

Approximately 33% of children were classified as suspected dengue cases and 43% of the children were anemic (median [IQR]: 111 [103, 118] g/L). Additionally, 26% of the cohort presented a fever and the most prevalent symptoms were rash (67%), nausea (43%) and vomiting (31%). Further, 7% of the children were stunted, 3% were underweight, and 5% of children were wasted. Approximately 19% were overweight and 12% were obese. Anemia was associated with an increased likelihood of presenting a rash (P = 0.04) overall as well as within the subset of children not suspected of having dengue viral infection. There were no significant associations noted between nutritional factors and other clinical variables.

Conclusions

Findings suggest that the prevalence of anemia and overweight were high in children presenting with suspected dengue virus infection in coastal Ecuador.

Funding Sources

NIH.

Reproductive History and Burden of Adverse Pregnancy Outcomes in Women in Southern India (P11-006-19)

Objectives

To characterize the reproductive history and pregnancy outcomes in women of reproductive age as part of an ongoing periconceptional surveillance program in Southern India.

Methods

Participants were women of reproductive age (15-40 y) who were not pregnant or lactating and resided in households within the 50 km2 catchment area of our community-based research site in Southern India (n = 813). After obtaining informed consent/assent, reproductive and obstetric histories were captured by a trained OB/GYN nurse via an electronic interviewer-administered questionnaire with the study participants.

Results

In this population, most participants reported being married (76.1%) and having some formal education (83.7%), with a median age of 30 (interquartile range [IQR] 24, 36) years. The average age of menarche reported was 13.0 (IQR 12.0, 14.0) years, and the median age at first pregnancy was 19.0 (IQR 17.0, 21.0) years. A total of 76.6% of women reported at least one previous pregnancy; 7.9% were primiparous and 65.7% were multiparous, with a median of 2.0 (IQR 2.0, 2.0) children currently living per woman. Approximately 80% of women reported taking any iron (80.4%) or folic acid (77.9%) supplements at any point during pregnancy. To date, women reported a total of 1,614 previous pregnancies with 1,318 live births (n = 1,308 singleton live births, n = 5 sets of twins), 169 spontaneous abortions, 94 induced abortions, and 42 stillbirths. Fifteen pregnancies were reported to have been affected by birth defects (n = 1 cleft lip, n = 2 cleft palate with cleft lip, n = 2 talipes equinovarus/clubfoot, n = 3 spina bifida, n = 3 encephalocele, n = 1 anencephaly, and n = 3 other birth defects). A total of 48 women reported a biological relative with a birth defect diagnosed at birth.

Conclusions

A substantial burden of birth defects was reported in this population (7 neural tube defects in 1,614 pregnancies). Findings from this survey will inform the establishment of an ongoing birth defects surveillance system and a randomized efficacy trial for prevention of anemia and birth defects in Southern India.

Funding Sources

Centers for Disease Control and Prevention.

AF was supported by the National Institutes of Health #5 T32 HD087137

Markers of Iron Status as Predictors, Effect Modifiers and Outcomes in an RCT of Growth Hormone and Rosiglitazone for Obesity and Insulin Resistance in Adults Living with HIV (P18-072-19)

Objectives

This secondary analysis was conducted to examine the association of iron status with changes in insulin sensitivity (SI, µU*10–4*min1*ml−1) and visceral adipose tissue (VAT, L) as part of a randomized, double blind, placebo-controlled trial. This analysis was conducted to examine if serum ferritin (µg/L) and hemoglobin (Hb, g/dL)·independently predict changes in SI or VAT·are affected by treatment with rosiglitazone or recombinant human growth hormone (rhGH)·modify the effect of changes in SI or VAT in response to treatment among adults living with HIV with abdominal obesity and insulin-resistance participating in a randomized trial.

Methods

In a 2 × 2 double-blinded factorial design, participants were provided 4 mg rosiglitazone twice daily, 3 mg rhGH daily, a combination of rhGH + rosiglitazone, or double-placebo (control) for 12 weeks. Serum ferritin was adjusted for inflammation (C-reactive protein, mg/L; α-1-acid glycoprotein, g/L) using the Thurnham method. Generalized linear models were used to assess the association of baseline Hb and ferritin with changes in SI and VAT. Models were adjusted for the a priori determined covariates, age, treatment group, CD4 T-cell count and HIV-1 RNA viral load. Effect modification was assessed by including an interaction term between biomarker and intervention group. Generalized estimating equations were used to assess the effects of treatment on Hb and serum ferritin concentrations.

Results

At baseline, there was no difference in serum ferritin (n = 47) or Hb (n = 45) across treatments; overall median (Q1, Q3) ferritin 95.60 (48.13, 151.00) µg/L, P = 0.13, Hb 14.40 (13.10,15.10) g/dL, P = 0.48. Neither marker of iron status was an independent predictor of changes in SI or in VAT. Compared to placebo at 12 weeks, Hb was significantly reduced by rosiglitazone (ß = −1.46, SE = 0.35, P < 0.0001), and by rhGH/rosiglitazone (ß = −0.93, SE = 0.30, P = 0.002) but not by rhGH alone. There was no significant effect of treatment on serum ferritin concentrations. Neither ferritin nor Hb modified the effect of intervention on changes in SI or VAT.

Conclusions

In this study among PLHIV, iron status was not an independent predictor or effect modifier of changes in SI or VAT. Treatments with rosiglitazone significantly reduced Hb concentration, which is consistent with adverse effects reported in previous clinical trials.

Funding Sources

Division of Nutritional Sciences, Cornell University.

Weill Cornell Medicine, Cornell University

Diet and the Gut Microbiome in 10–18-Month-Old Children Living in Urban Slums of Mumbai, India (OR01-08-19)

Objectives

1. To characterize the gut microbiome and 2. explore associations between dietary intake and gut microbiome composition and predicted function among children at screening for participation in a randomized controlled feeding trial in urban slums of Mumbai, India.

Methods

Young children (10–18 months old) living in urban slums of Mumbai were screened from March to November 2017 after obtaining informed consent from caregivers. Nutrient intakes from 24-hour dietary recall were analyzed as absolute intakes, % of RDA, and using the nutrient residual to adjust for energy intake. DNA extracted from rectal swabs was sequenced (16S rRNA V3-V4 region; Illumina MiSeq) and further processed using QIIME2 and PICRUSt algorithm. Linear regression was used to estimate associations between nutrient intakes (adjusted for age and sex) and measures of gut microbiome α-diversity, including Shannon Index, Chao1 estimator, observed OTUs, Proteobacteria relative abundance, Firmicutes: Bacteroidetes ratio, and predicted KEGG pathways. Statistical analysis was performed using RStudio, and SAS 9.4.

Results

A subset of 53 children with rectal swabs collected were included. Sequencing yielded 8984,126 reads including 2173 unique operational taxonomic units (OTUs). Mean relative abundance of Proteobacteria was 78.2%. Shannon Diversity Index was positively associated with absolute intakes of total carotenoids and beta-carotene. Children with adequate intakes of zinc and beta-carotene had a higher mean Shannon Diversity Index. The Chao1 estimator was positively associated with absolute intakes of calcium. The number of observed OTUs was positively associated with calcium and zinc intakes. The predicted gene counts for germination, sporulation, amoebiasis, and protein digestion and absorption pathways were also associated with both macro- and micro-nutrient intakes.

Conclusions

The gut microbiome in this sample of young children in Mumbai was dominated by Proteobacteria, which includes many potentially pathogenic species. Dietary intakes of macro- and micronutrients were associated with measures of diversity and predicted functional pathways after adjustment for age and sex, warranting further longitudinal study of the diet's impact on the microbiome particularly in young children.

Funding Sources

HarvestPlus.

Comparison of Inflammation Adjustment Strategies for Retinol-binding Protein in Acute and Convalescent Serum of Patients with Acute Febrile Illness in Guayaquil, Ecuador (P10-095-19)

Objectives

1. Assess serum retinol-binding protein (RBP) among clinical participants at two time points: during the onset of acute febrile illness and again during the convalescent recovery phase.

2. Apply and compare several inflammation adjustment strategies for determining vitamin A status.

Methods

Patients presenting with acute febrile illness were enrolled. Participants were asked to return for a follow-up convalescent visit 2–4 weeks later. Comprehensive demographic and clinical information along with blood samples were collected at both visits. RBP, C-reactive protein (CRP), and alpha-1-acid glycoprotein (AGP) were measured by ELISA, (R&D Systems, Inc., Minneapolis, MN), and pro-inflammatory cytokines were measured by magnetic bead multiplex assay, (EMD Millipore Corporation, Billerica, MA). Continuous RBP was adjusted using the Thurnham correction factor (TCF), the BRINDA regression correction (BRC), and the CRP-only adjustment factor (CAF). An interleukin 6 (IL-6) regression model was also applied to adjust RBP based on IL-6 concentrations for participants with CRP ≥5 mg/L. Adjusted RBP concentrations were compared between paired visits using non-parametric paired Wilcoxon signed-rank tests, and p values <0.05 were considered significant.

Results

Blood samples were collected from 18 participants during acute and convalescent phase visits. The unadjusted median (quartile 1-quartile 3) RBP concentration (µmol/L) during the acute visit, 1.02 (0.90–1.21), was lower than the convalescent visit, 1.57 (1.26–1.75), P = 0.0004. After applying the TCF, BRC, and CAF, median RBP between visits differed significantly, similar to the unadjusted RBP. RBP adjusted using the IL-6 regression model was not significantly different between paired visits 1 and 2, 1.40 (1.11–1.76) and 1.62 (1.26–1.97) respectively, P = 0.12.

Conclusions

Circulating unadjusted RBP concentrations were significantly lower in participants during the acute illness phase compared to the convalescent recovery phase. Under the assumption that underlying vitamin A status remains constant during a 2–4 week period, adjusting RBP for IL-6 in participants with CRP ≥5 mg/L was the only strategy in which adjusted RBP did not differ significantly between paired visits.

Funding Sources

NIH National Institute of Biomedical Imaging and Bioengineering, NIH Office of Research on Women's Health.

Biomarkers of Heavy Metal Exposure in Women of Reproductive Age in Southern India (P24-054-19)

Objectives

To examine biomarkers of heavy metal exposure in women of reproductive age as part of an ongoing periconceptional surveillance program in Southern India.

Methods

Participants were women of reproductive age (15–40 y) who were not pregnant or lactating and resided in households within the 50 km2 catchment area of our community-based research site. Venous blood samples were collected in metal-free blue top K2EDTA vacutainers, and urine samples were collected in acid-washed containers. Plasma and red blood cells were separated by centrifugation, processed, and stored <−80 °C until analysis. Heavy metal concentrations of lead, mercury, arsenic, cadmium, and chromium were measured via inductively coupled plasma-mass spectrometry in a pilot sub-sample (n = 70). Descriptive statistics for plasma heavy metal concentrations were reported (median, interquartile range [IQR]).

Results

Half of participants had at least one heavy metal detected in plasma (48.6%). For lead, 30.0% of women had detectable concentrations in plasma (median (IQR): 0.2 (0.2, 0.2) µg/dL), with 1.4% having plasma lead concentrations ≥0.5 µg/dL. For mercury, 17.1% of participants had detectable concentrations in plasma (median (IQR): 4.5 (2.0, 8.0) µg/L) and 7.1% of samples were >5.8 µg/L. A total of 31.1% of participants had detectable chromium plasma concentrations (median (IQR): 2.5 (1.0, 15.0) µg/L), with 15.7% of samples >2.1 µg/L. Concentrations of two or more heavy metals detected in plasma were found in 41.4% of participants: 7.1% lead and mercury, 18.6% lead and chromium, 10.0% mercury and chromium, and 5.7% with all three detected in plasma. No participants in this pilot sub-sample had plasma concentrations of arsenic or cadmium above the detection limit for this method (≥1.0 µg/L).

Conclusions

Findings from this pilot study provide preliminary evidence of heavy metal exposure – including lead, mercury, and chromium – in women of reproductive age in this setting. Because of potential interactions between exposure to heavy metals and folate status that may impact the risk of birth defects, biomarker surveillance of heavy metals remains a critical component of this program.

Funding Sources

Centers for Disease Control and Prevention.

AF was supported by the National Institutes of Health #5 T32 HD087137

Nutrient Intake and Adequacy Among Children 12–18 Months of Age in the Urban Slums of Mumbai, India (P11-093-19)

Objectives

a) To examine the mean nutrient intake among young children 12–18 months of age in urban slums of Mumbai, b) to determine the adequacy of micronutrient intake in their diet as per the National Institute of Nutrition (NIN) recommendations for optimum growth and development, and c) to determine if there is a difference in the nutrient intake of those randomized vs not randomized to the trial involving feeding of iron- and zinc-biofortified pearl millet complementary foods.

Methods

A total of 407 children were screened of which dietary data was available for 359 children. Of the total randomized (n = 223), dietary data was available for 216 children who were randomized to one of the two study arms (Arm 1: n = 107, Arm 2: n = 109). A 24-hour dietary recall was performed, and nutrient intake was calculated using CS dietary software using the NIN food composition tables.

Results

The average age of children was 14.81 months; 51% were males and 49% females. Majority of the children (81.64%) breastfed the previous day. The mean intake per day of calories, protein, iron, and zinc from complementary foods was found to be 598 kcal, 19.9 g, 3.25 mg, and 1.85 mg, respectively. On average, complementary feeding contributed to 89.6% of RDA for energy, 82.4% of calcium, 57.9% of iron, 58.8% of zinc, 51.9% vitamin A, 7.2% of vitamin B12, and 83.3% of dietary folate. All children had adequate caloric and protein intake. However, > 90% of children consumed inadequate iron, zinc, and folate according to the nutrient density of the complementary food, and > 50% of children consumed inadequate vitamin A and calcium. Overall, children randomized to the intervention were found to have a significantly higher intake of calories, protein, folate, and zinc.

Conclusions

Majority of children living in the urban slums of Mumbai consumed diet inadequate for most micronutrients. Our findings confirm the suitability of this population for interventions aiming to improve micronutrient intake

Funding Sources

HarvestPlus.

Vitamin B12 in HIV-Infected Patients with Insulin Resistance and Visceral Adiposity in a Randomized Trial of Recombinant Human Growth Hormone and Rosiglitazone (OR12-05-19)

Objectives

To 1) examine the burden of vitamin B12 deficiency, 2) determine the effects of recombinant growth hormone and rosiglitazone on vitamin B12 status, and 3) investigate vitamin B12 status as a risk factor for changes in insulin sensitivity (SI) and visceral adipose tissue (VAT), among HIV-infected patients with insulin resistance and visceral adiposity participating in a randomized controlled trial.

Methods

Participants were 72 HIV-infected adults (median [IQR]: 48 [43, 53] y) with insulin resistance and visceral adiposity who participated in a 12-week randomized trial of recombinant growth hormone (GH) and rosiglitazone (R). Venous blood samples were collected at baseline, 4, and 12 weeks, and samples were centrifuged, processed, and stored <−80°C until analysis. Total vitamin B12, methylmalonic acid (MMA), homocysteine, and serum folate were measured in a subset. Vitamin B12 deficiency and insufficiency were defined as <148.0 pmol/L and <221.0 pmol/L, respectively. Elevated MMA was defined as >0.26 µmol/L.Generalized estimating equations were used to evaluate the effects of treatment on vitamin B12 status. Generalized linear models were used to assess the associations of vitamin B12 concentrations with SI and VAT.

Results

A total of 2.3% of patients were vitamin B12 deficient and 13.6% were vitamin B12insufficient at baseline (median [IQR]: 419.8 [287.0, 538.6] pmol/L); 5.4% had elevated MMA concentrations. The GH + R intervention significantly lowered vitamin B12 concentrations at 4 (β: −66.6, 95% CI: −119.6, −13.6, P = 0.01) and 12 (β: −73.3, 95% CI: −117.8, −28.7, P = 0.001) weeks, compared to placebo. The GH-only intervention significantly lowered vitamin B12concentrations at 12 weeks (β: −89.1, 95% CI: −143.0, −35.3, P = 0.001), compared to placebo. Vitamin B12 concentrations did not significantly change in the R-only group. There were no significant changes in MMA concentrations. Vitamin B12 concentrations at baseline were not significantly associated with changes in SI or VAT (P > 0.05).

Conclusions

The prevalence of vitamin B12 deficiency was low in HIV-infected individuals with insulin resistance and visceral adiposity. However, interventions containing recombinant growth hormone decreased vitamin B12 status during follow-up.

Funding Sources

BG was supported by NIH/NCATS Grant # TL1-TR-002386; Division of Nutritional Sciences, Cornell University.

Does cranberry have a role in catheter-associated urinary tract infections?

INTRODUCTION

Catheter-associated urinary tract infections (CA-UTIs) are a prevalent and costly condition, with very few therapeutic options. We sought to evaluate the efficacy of an oral cranberry supplement on CA-UTIs over a six-month period.

METHODS

Subjects with long-term indwelling catheters and recurrent symptomatic CA-UTIs were enrolled to take a once-daily oral cranberry supplement with 36 mg of the active ingredient proanthocyanidin (PACs). Primary outcome was reducing the number of symptomatic CA-UTIs. This was defined by ≥10³ (cfu)/mL of ≥1 bacterial species in a single catheter urine specimen and signs and symptoms compatible with CA-UTI. Secondary outcomes included bacterial counts and resistance patterns to antibiotics.

RESULTS

Thirty-four patients were enrolled in the trial; 22 patients (mean age 77.22 years, 77.27% were men) completed the study. Cranberry was effective in reducing the number of symptomatic CA-UTIs in all patients (n=22). Resistance to antibiotics was reduced by 28%. Furthermore, colony counts were reduced by 58.65%. No subjects had adverse events while taking cranberry.

CONCLUSIONS

The cranberry supplement reduced the number of symptomatic CA-UTIs, antibiotic resistances, and major causative organisms in this cohort. Larger, placebo-controlled studies are needed to further define the role of cranberry in CA-UTIs.

Surveillance of pneumococcal colonization and invasive pneumococcal disease reveals shift in prevalent carriage serotypes in Massachusetts' children to relatively low invasiveness

BACKGROUND

Following the introduction of pneumococcal conjugate vaccines (PCV), overall nasopharyngeal colonization rates have not changed significantly, however a dramatic and sustained decline in invasive pneumococcal disease (IPD) in children was observed in every setting where the PCVs were implemented. We aimed to describe the differences in invasive disease potential of serotypes that are common colonizers in pre- and post-vaccine eras in order to provide further insight in our understanding of dynamic epidemiology of pneumococcal diseases.

METHODS

Using data from surveillance of nasopharyngeal carriage and enhanced surveillance for IPD, a serotype specific "invasive capacity (IC)" was computed by dividing the incidence of IPD due to serotype x by the carriage prevalence of that same serotype in children <7years of age in Massachusetts. We have evaluated the serotype specific invasive capacity in two periods; pre-PCV13 (2001/02, 2003/04, 2006/07, 2008/09) and post-PCV13 (2010/11 and 2013/14), and by age groups; <24monthsvs. ≥24months.

RESULTS

An approximate 50-fold variation in the point estimate was observed between the serotypes having the highest (7F, 38, 19A, 3, 33F) and the lowest (6C, 35B, 21, 11A, 23B and 23A) computed serotype specific invasive disease potential. In the post-PCV13 era (6C, 35B, 11A, 23B and 23A), 5 of the 7 most common serotypes colonizing the nasopharynx were serotypes with the lowest invasive capacity. Serotype specific invasive capacity trended down in older children for majority of the serotypes, and serotypes 3, 10A and 19A had significantly lower invasive capacity in children older than 24months of age compared to younger children.

CONCLUSION

Invasive capacity differs among serotypes and likely by age. Point estimates of IC for most of the common serotypes colonizing children in Massachusetts in post-PCV13 era were low and likely explain the continued reduction in IPD from the pre-PCV era in the absence of specific protection against these serotypes.

Improving value-add work and satisfaction in medical residents training: a resident-led quality improvement project employing the lean method to improve hospital supply usage

PURPOSE

Our hospital has a Housestaff Quality Council that fosters education and mentorship of medical residents for quality improvement methodologies. Medical residents on our council identified non-standardised storage rooms as a source of medical resident inefficiency and dissatisfaction. To improve value-add work, medical residents implemented and evaluated a quality improvement project of storage room supplies using the lean method.

METHODS

Using 5S principle and lean methodology, we designed and implemented a standardised supply cart with physician specific supplies. Between April 2014 and April 2015, 40 random observations (20 residents and 20 nurses) both before and after the standardised supply cart implementation were made. The duration time to locate an item was measured in seconds. The paths taken to locate items were drawn as spaghetti diagrams. Nurses served as our control group given that their supplies were not moved in the implementation. Fifty residents were surveyed to assess their satisfaction.

RESULTS

Implementation of the standardised supply cart reduced the time for residents to locate an item per visit from 50.8 to 30.2 s in one unit (p<0.05) and 127 to 28.3 s in the second unit (p<0.05). Mean time savings per day per resident were 5 min. The spaghetti diagrams indicated that finding supplies became more efficient after the intervention for residents. After the intervention, 92% of residents reported finding supplies more rapidly and 86% reported less frustration with finding supplies.

CONCLUSIONS

Residents applied the 5S principles and lean methodology to identify and solve a problem that created inefficiency and dissatisfaction.

Inter-rater reliability between musculoskeletal radiologists and orthopedic surgeons on computed tomography imaging features of spinal metastases

INTRODUCTION

The primary objective of this pilot study was to examine the inter-rater reliability in scoring the computed tomography (ct) imaging features of spinal metastases in patients referred for radiotherapy (rt) for bone pain.

METHODS

In a retrospective review, 3 musculoskeletal radiologists and 2 orthopedic spinal surgeons independently evaluated ct imaging features for 41 patients with spinal metastases treated with rt in an outpatient radiation clinic from January 2007 to October 2008. The evaluation used spinal assessment criteria that had been developed in-house, with reference to osseous and soft tissue tumour extent,presence of a pathologic fracture,severity of vertebral height loss, andpresence of kyphosis.The Cohen kappa coefficient between the two specialties was calculated.

RESULTS

Mean patient age was 69.2 years (30 men, 11 women). The mean total daily oral morphine equivalent was 73.4 mg. Treatment dose-fractionation schedules included 8 Gy/1 (n = 28), 20 Gy/5 (n = 12), and 20 Gy/8 (n = 1). Areas of moderate agreement in identifying the ct imaging appearance of spinal metastasis included extent of vertebral body involvement (κ = 0.48) and soft-tissue component (κ = 0.59). Areas of fair agreement included extent of pedicle involvement (κ = 0.28), extent of lamina involvement (κ = 0.35), and presence of pathologic fracture (κ = 0.20). Areas of poor agreement included nerve-root compression (κ = 0.14) and vertebral body height loss (κ = 0.19).

CONCLUSIONS

The range of agreement between musculoskeletal radiologists and orthopedic surgeons for most spinal assessment criteria is moderate to poor. A consensus for managing challenging vertebral injuries secondary to spinal metastases needs to be established so as to best triage patients to the most appropriate therapeutic modality.

The impact of regional anesthesia on bladder cancer outcomes

e15021

Background: Studies in breast and prostate surgery show reduced cancer recurrence after regional (RA) versus general anesthesia (GA). Mechanisms include RAÕs reduced post-operative opioid use and cortisol-mediated immunosuppression. RA may be used alone during transurethral resection of bladder tumors (TURBT) and in combination with GA during radical cystectomy (RC). We assess the impact of RA on short-term bladder urothelial cell carcinoma (UCC) recurrence after TURBT or RC. Methods: From 8/2001 and 6/2006 to 6/2011, 151 patients underwent RC and 488 patients underwent TURBT for bladder UCC, respectively. Those with incomplete resection on TURBT were excluded. Anesthesia included RA or GA for TURBT, and GA alone or GA + RA for RC. Multivariate logistic regression was performed to identify significant predictors of biopsy- or radiography-confirmed UCC recurrence. Results: TURBT. Of 252 patients, 211 received GA and 41 received RA during TURBT. Patient and operative characteristics were similar between groups. Recurrence was 56% at 12 months for GA and RA. Multivariate analysis revealed clinical stage to be the only predictor of UCC recurrence (HR=1.8, p<0.0001). Anesthesia had no affect on 6 or 12 month RFS, DSS or OS (see table). RC. GA was used in 114 patients and 37 patients had GA + RA at RC. There were no between group differences in patient or tumor characteristics. After follow-up of 18 months, 25.9% and 21.6% recurred in GA and GA+RA groups, respectively (p>0.05). There were no differences in RFS, DSS, or OS (see Table). Conclusions: Contrary to other malignancies, our data suggest anesthesia type at TURBT or RC does not affect bladder cancer outcomes. Anesthesia modality should be based on patient comorbidities and procedure type. [Table: see text]

Open Versus Laparoscopic Versus Robot-Assisted Laparoscopic Prostatectomy: The European and US Experience

Open radical prostatectomy (ORP) is the reference standard for the surgical management of localized prostate cancer. With wider availability of minimally invasive radical prostatectomy techniques, there is a debate regarding the standard treatment of the management of localized prostate cancer. Therefore, we reviewed the current status of laparoscopic radical prostatectomy (LRP) and robotic-assisted laparoscopic radical prostatectomy (RALRP) as compared with ORP. Because no prospective, randomized trials comparing the different techniques have been performed, outcomes must be assessed from published series by centers that focus on ORP, LRP, and RALRP. Aside from reducing the amount of blood loss, current data suggest that the most significant outcomes (cure, continence, and potency) are no better with LRP or RALRP than with conventional ORP. Therefore, in experienced hands, ORP remains the gold standard procedure. However, there is a trend toward consistently better outcomes following RALRP in comparison with LRP. In the end, individual patient outcomes can be maximized by choosing the best modality based on the patient's comorbid medical conditions, cancer characteristics, and surgeon experience. Future studies are needed to further investigate long-term cancer control as well as functional outcomes for RALRP series.