What Is Known/What Is New
What Is Known
- Children with severe acute malnutrition have morphological changes in the gut characterized by decreased mucosal thickness, increased permeability, and increased infiltration of the lamina propria with inflammatory cells.
What Is New
- Reduced plasma citrulline and increased fecal myeloperoxidase and neopterin biomarkers support evidence of reduced enterocyte mass and increased intestinal inflammation in children with severe acute malnutrition.
- Children with diarrhoea, edema, HIV, dermatosis, C-reactive protein >10 g/dL have greater reduction in the enterocyte mass.
- The enterocyte damage is likely to impair nutrient absorption and increase the risk of infections.
Severe acute malnutrition (SAM) affects nearly 19 million children under 5 years worldwide (1). Mortality among these children is associated with systemic and intestinal inflammation (2,3). Biopsy studies revealed significant morphological changes in the intestinal mucosa of children with SAM characterized by decreased mucosal thickness, shortened villi, and increased inflammatory cells in the lamina propria (4–6). Such characteristics have been described in relation to a defective absorptive surface and compromised intestinal barrier function. Although intestinal biopsies are informative, sampling requires invasive procedures. Simple, reliable, and noninvasive methods are needed to assess the intestinal function.
Blood and fecal biomarkers are used for evaluation of intestinal function in several diseases. Plasma citrulline, produced almost exclusively by the enterocytes, is a marker of enterocyte mass in inflammatory bowel diseases, HIV patients, and critically ill patients (7–12). However, its application in the context of children with SAM is yet to be explored. Quantitative real-time PCR has been used to detect bacterial and fungal DNA in blood (13–15). Presence of bacterial and fungal DNA in blood may reflect bacterial translocation from the gut because of reduced intestinal integrity. Fecal myeloperoxidase (MPO) and fecal neopterin (NEO) have been used as markers of intestinal inflammation (16–21). Fecal MPO and NEO can be sampled using limited resources in the field, are relatively stable compounds, and are resistant to proteolysis.
We describe gut function of children hospitalized with complicated SAM using blood and fecal biomarkers and the correlates of impaired gut function.
Study Design and Setting
The study was cross-sectional, nested in a randomized trial (www.isrctn.com, ISRCTN 16454889) among 400 children with SAM conducted at Mwanamugimu Nutrition Unit (MNU), Mulago National Referral Hospital in Kampala, Uganda from March 2014 to July 2015. Patients were managed based on the Ugandan protocol for Integrated Management of Acute Malnutrition (IMAM) (22). Children were first screened for SAM at the paediatric emergency unit. SAM was defined as weight-for-height z score (WHZ) < −3SD or MUAC <11.5 cm or presence of bilateral pitting edema. Children with complicated SAM were admitted and transferred to MNU for treatment. At MNU, the study medical officer re-assessed the patients and selected those that met the eligibility criteria. The study included children ages 6 to 59 months with SAM and those for whom caregivers provided an informed consent. Children with body weight less than 4.0 kg at admission or patients with severe conditions, such as shock or severe respiratory distress at admission or significant congenital diseases or malignant diseases were excluded. Children ages 6 to 59 months with WHZ >−1 were enrolled as controls.
Ethical approval was obtained from Makerere University, School of Medicine Research, and Ethics Committee (REF# REC 2013 132) and a consultative approval was given by The National Committee of Health Research Ethics in Denmark. Caregivers provided a written informed consent before enrollment into the study.
A questionnaire administered by a trained medical officer was used to collect data on demographic characteristics, maternal age, and level of education. Caregiver-reported symptoms of the child were noted and caregivers were also asked to rate the severity of illness of his/her child during the 2 weeks before admission on a visual analogue scale (VAS) of 0 to 10. The visual analogue scale consisted of pictograms of faces describing perceived appearance of the child. The study clinician performed a standardized physical examination of the child.
Body weight was measured using a digital weight scale (Hamburg, Germany), length or height using an infant length board (Weight and Measure LLC, Annapolis, MD) (22) and MUAC using a colour-coded tape. WHZ and height-for-age z scores (HAZ) were computed using WHO Anthro version 3.2.2.
Approximately 4 ml of venous blood was collected into heparinized evacuated tubes (Becton Dickinson, Franklin Lakes, NJ). One milliliter was used to determine the blood cell counts and HIV sero-status. Determine HIV-1/2 rapid tests (Abbort Laboratories, Chicago, IL) and HIV 1/2 Stat-Pak Dipstick Assay kit (Abbort Laboratories) were used to evaluate HIV sero-status for children above 18 months whereas HIV DNA PCR test was used to confirm the HIV status of children less than 18 months at Baylor HIV clinic. A drop of whole blood (40 μl) was placed on standard Guthrie filter paper (Sartorius Stedium Biotech, Pekema, Denmark) and allowed to air-dry for 36 hours. The dried blood spots (DBS) were stored at −20°C in a zip lock plastic bag and shipped on dry ice to University of Copenhagen (UCPH) for analysis of the bacterial 16S rRNA gene and Candida-specific ITS fragment. Whole blood was centrifuged at 1300 to 2200g for 10 min and the supernatant pipetted into separate cryo vials, samples stored at −80°C and shipped on dry ice to UCPH for analysis of C-reactive protein (CRP) and plasma citrulline.
Approximately 10 g of stool was collected, emptied into a 2 ml cryo tube and stored at -80°C. Samples were shipped on dry ice to UCPH for analysis of fecal MPO and NEO.
Analysis of C-reactive Protein and Gut Function Biomarkers
CRP was analyzed by a high-sensitivity latex-improved immunoturbidity assay (Horiba, France).
Detection of plasma citrulline by liquid chromatography mass spectrometry (LC-MS) was performed as described elsewhere (23). Samples were processed with Sirocco protein precipitation plate (Waters, Milford, MA) and separated on an ACQUITY UPLC HSS T3 Column (Waters) on an ultra-performance liquid chromatography in tandem with triple quadrupole detector mass spectrometry. l-Citrulline-4, 4, 5, 5-d4 (Sigma-Aldrich, St. Louis, MO) was used as an internal standard. Quantification of citrulline was carried out using QuanLynx (Waters).
Approximately, 0.33 g stool was measured out using an analytical weight scale and placed in a glass tube. Three milliliters of PBS 1% Tween20 was pipetted into the glass tube, centrifuged at 3000g for 20 minutes and supernatant pipetted into a 2 mL Eppendorf tube and further centrifuged at 13 000g in 5 minutes. Different dilutions were used based on the standard curves. Most samples for fecal MPO were diluted at 1:20 000 and analyzed using Duoset ELISA Human Myeloperoxidase (R&D systems, Mineapolis, MN). For NEO, dilutions of 1:100 were performed and a ready-to-use ELISA kit RE59321 Neopterine Elisa, IBL (Salzburg, Austria) was used for analysis.
Seven 3 mm diameter punches from each dry blood spot (DBS) on the Guthrie filter papers were cut with Uni-Core Punch (GE Healthcare, Chicago, IL) and ejected into an Eppendorf tube. Cells were lyzed with 300 μL Cell Lysis Solution (Qiagen, Hilden, Germany) and further processed by the Gentra Pure Gene Tissue Kit (Qiagen). DNA was eluted with 50 μL DNA hydration solution and was analyzed by high-throughput qPCR using a BioMark HD System (Fluidigm, San Francisco, CA) essentially as described by the supplier (see Supplemental Text, Supplemental Digital Content, http://links.lww.com/MPG/B648, for detailed description of analysis using the BioMark HD System). Each sample was run in duplicate and mean values of 16S rRNA gene and ITS fragment copies were used for data analysis. To assess contamination, if any, 7 3 mm diameter filter paper spots (away from the DBS spots) were cut from random DBS cards and analyzed as described above. Nontemplate controls (negative controls) were also included for assessment of carryover contamination.
Data was analyzed using Stata version 12 (Statacorp, College Station, TX). Chi-square test and the Students t-test were used to test for differences between children with SAM and the controls for normally distributed variables while Mann-Whitney (Wilcoxon rank sum) test was used to compare medians for nonnormally distributed variables. Plasma citrulline, fecal MPO, fecal NEO, bacterial 16S rRNA gene, and Candida-specific ITS fragment copy numbers were not normally distributed hence data was log transformed and linear regression analysis performed while adjusting for age and sex.
A total of 400 children with SAM and 30 community controls were enrolled into the study. The children with SAM were on average 9 months younger than the controls (17 vs 26 months, P < 0.001) (Table 1). More than half of the children with SAM showed plasma CRP greater than 10 mg/L (62%) and reported symptoms of cough (66%), diarrhea (61%), and bilateral pitting edema (66%). Twenty-seven (27) children presented with dermatosis, of which 93% had grade 3 edema. The HIV prevalence was 11% among children with SAM (Table 1).
A total of 295 samples from children with SAM and 27 controls were used for analysis of plasma citrulline Median plasma citrulline was lower among children with SAM compared with controls (5.14 vs 27.4 μmol/L, P < 0.001) (Table 2). Among the children with SAM, edema, diarrhoea, dermatosis, severity of illness as rated by the caregiver and plasma CRP >10 mg/L correlated negatively with plasma citrulline all of which remained negative correlates of plasma citrulline in a mutually adjusted model (Table 3). See Supplemental Figure (Supplemental Digital Content, http://links.lww.com/MPG/B648), scatter plot showing distribution of the blood and fecal biomarkers among children with SAM with or without diarrhoea.
A total of 349 and 365 stool samples were available for MPO and NEO analysis respectively. Median fecal MPO was higher among children with SAM compared with controls (18083 vs 7482 ng/mL, P = 0.001) (Table 2). Among the children with SAM, WHZ showed negative correlation with fecal MPO corresponding to a 13% lower fecal MPO per unit increase in WHZ.
Median fecal NEO was also higher among children with SAM compared with controls (541 vs 210 nmol/L, P < 0.001) (Table 2). Among the children with SAM, elevated neutrophil and monocyte counts and symptoms of fever and cough showed positive correlation with fecal NEO, whereas age, MUAC, WHZ score were negative correlates (Table 4). Children with edema generally had lower fecal NEO. For example, children with grade 2 and 3 edema had 36 and 53% lower mean fecal NEO levels, respectively.
A total of 345 samples of children with SAM and 27 controls were used for qPCR-analysis of bacterial and Candida DNA in the blood. Children with SAM appeared to have a higher median blood bacterial 16S rRNA gene than the controls, although the difference was only marginally significant (95 [19:303] vs 28 [11:133] copies/μL, P = 0.05) (Table 2). Children with SAM did not have significantly higher median Candia-specific ITS fragment copies in their blood compared with the controls (33 [8:120] vs 24 [4:115]; P = 0.46) (Table 2). Nevertheless, the level of ITS fragment copies among children with SAM was 1.71 (95% confidence interval [CI] 1.04–2.79, P = 0.03) times higher in those with plasma CRP >10 mg/L compared with those with plasma CRP <10 mg/L (see Supplemental Table, Supplemental Digital Content, http://links.lww.com/MPG/B648 that shows the correlates of 16S rRNA gene and ITS fragment copies in blood).
Mean values obtained for areas of the filter papers without DBS were 8.29 (16S rRNA gene) and 2.10 (Candida-specific ITS fragment), indicating that prior contamination of the filter paper does not significantly contribute to the overall DNA analyzed and that the vast majority of the DNA analyzed represents what was actually present in the DBS spots. Likewise, mean values obtained for negative controls were 2.37 ± 2.68 (16S rRNA gene) and 3.87 ± 5.98 (Candida-specific ITS fragment), ruling out the possibility of carry over contamination.
Our findings suggest that gut function is impaired among children with complicated SAM. Impaired gut function was characterized by increased intestinal inflammation, increased microbial translocation, and reduced enterocyte mass.
We found very low plasma citrulline levels among children with SAM compared with the controls, whose plasma citrulline was largely within the normal range of 20 to 40 μmol/L (11). As citrulline is produced almost exclusively by enterocytes, studies among patients with inflammatory bowel disease and HIV-associated enteropathy have used plasma citrulline as a biomarker of enterocyte mass (9–11). As far as we are aware, this is the first article to explore the use of plasma citrulline as a marker of enterocyte mass among children with SAM. It is plausible that the low plasma citrulline levels found in our study is because of a reduced enterocyte mass, because biopsy-based studies among children with SAM have showed a markedly damaged intestinal epithelium (5,24). Under conditions of malnutrition, it is also possible that alterations in metabolic pathways involved in the synthesis of citrulline in the enterocytes occur. Semba et al (25) found low-serum citrulline, tryptophan, ornithine, phosphatidylcholines, and sphingomyelins levels among Malawian children with increased intestinal permeability and EED. They attributed the low-serum metabolite levels to alterations in metabolic pathways, such as citrulline synthesis, tryptophan-serotonin pathway, and the Kynurenine pathway (25). Further evaluation of amino acid markers coupled with citrulline metabolism in children with SAM is needed.
We found an inverse correlation between plasma citrulline and elevated CRP, an established marker of systemic inflammation. Kosek et al (26) found this association between citrulline and CRP and other markers of systemic inflammation in a longitudinal study among children with EED. Diarrhea was associated with low-plasma citrulline in this study consistent with the findings of a study in Zambia where plasma citrulline was significantly lower among individuals with HIV and enteropathy who recently had diarrhea (7). This is plausible as intestinal biopsies among children with SAM have shown decreased mucosal thickness in the presence of diarrhea (6,27).
Bilateral pitting edema was a negative correlate of plasma citrulline in our study. Biopsy studies have also suggested that the intestinal mucosa was more affected in children with edematous SAM compared with nonedematous SAM (5,24). The explanations for this are not clearly known. The low plasma citrulline in the children with edematous SAM may also be because of the alterations in the body fluid compartments because of a general dilution of the extracellular (fluid) space with edema. We did not measure serum albumin levels in this study as some studies have shown no correlations between circulating plasma citrulline and serum albumin levels or prealbumin (28). Furthermore, as reviewed by Golden et al (29), low protein intake or low serum albumin are not responsible for edema or other manifestations of SAM. In line with this, the concentration of other free amino acids have also been found to be lower with increasing edema (30).
Fecal MPO and NEO were nearly 3 times higher in the children with SAM compared with controls. Studies among malnourished children have demonstrated presence of intestinal pathogens contributing to intestinal inflammation, as well as elevated markers of systemic inflammation (2,31). A recent study in Malawi found that 84% of malnourished children had more than 1 intestinal pathogen and 44% had multiple (>4) intestinal pathogens (2). The presence of intestinal inflammation may be associated with altered biodiversity of the gut microbiota (GM) (32). In line with this, a study conducted in Uganda among children with complicated SAM showed a high abundance of microorganisms from the Enterobacteriaceae family and a low GM biodiversity in those with nonedematous SAM relative to edematous SAM (32,33).
Children with edematous SAM had lower fecal NEO, and so did children with dermatosis, which persisted after controlling for edema. The skin manifestations in SAM are poorly understood, but are commonly seen among children with edematous SAM, and rarely among those with nonedematous SAM. The low fecal NEO could reflect less efficient cellular immune response in edematous children or it could simply reflect the previously reported trend for nonedematous children in hospital to have higher prevalence of symptomatic infections. Findings suggestive of an impaired immune response have been described in tuberculin skin tests often accompanied by false negative results in edematous SAM (34). Among the controls, fecal MPO and NEO levels were relatively high when compared with similarly aged well-nourished children in high-income countries, but similar to nonmalnourished healthy children living in poor environmental conditions in low- and middle-income countries (LMIC) (19). This is in line with studies suggesting the presence of EED in seemingly healthy children conducted in low-income countries (18,33).
We found a marginal difference in blood 16S rRNA gene count among children with SAM compared with the controls indicating increased bacterial translocation over the gut epithelium in SAM children possibly because of impaired intestinal barrier integrity. This is consistent with a study in Zambia, where elevated markers of bacterial translocation were found among Zambian children with SAM and adults compared with the apparently healthy children (6). We did not find a difference in Candida-specific ITS fragment copies among children with SAM and controls; however, among the children with SAM, plasma ITS fragment copies were positively correlated with elevated plasma CRP. The reasons for the relatively high numbers of 16S rRNA and ITS fragment copies in blood among the healthy controls remains obscure; however, a study has found similar findings among healthy subjects (14).
Our study has some limitations. Firstly, plasma citrulline has not been validated in children with SAM as a marker of enterocyte mass. Secondly, the ages of the children with SAM and controls were different; however, age was adjusted for during analysis. Finally, there was a possibility of overestimating bacterial and fungal DNA in blood as the DBS were not sterile and were dried in room air for several hours. We used a highly sensitive and specific real-time PCR test and our assessment of contamination showed 16S rRNA gene and ITS fragment on the filter papers without blood on to be below detection limits. We cannot, however, fully rule out that bacteria and fungi may have had better conditions for growth when the paper was soaked in blood.
In conclusion, children with complicated SAM have an impaired gut function characterized by increased intestinal inflammation, increased bacterial translocation and reduced enterocyte mass. This is likely to result in impaired nutrient absorption, increased risk of infection, and changes of gut microbiota composition. Further evaluations are needed to identify biomarkers of gut function in children with SAM, taking into consideration the systemic physiological changes that occur because of reductive adaptation in SAM.
1. Black RE, Victora CG, Walker SP, et al. Maternal and Child Nutrition Study Group. Maternal and child undernutrition and overweight in low-income and middle-income countries. Lancet
2. Attia S, Versloot C, Voskuijl W, et al. Mortality in children with complicated severe acute malnutrition is related to intestinal and systemic inflammation. Am J Clin Nutr
3. Jones KDJ, Berkley JA. Severe acute malnutrition and infection. Paediatr Int Child Health
4. Stanfield JP, Hutt MS, Tunnicliffe R. Intestinal biopsy in kwashiorkor. Lancet Lond Engl
5. Brunser O, Reid A, Monckeberg F, et al. Jejunal biopsies in infant malnutrition: with special reference to mitotic index. Pediatrics
6. Amadi B, Besa E, Zyambo K, et al. Impaired barrier function and autoantibody generation in malnutrition enteropathy in Zambia. EBioMedicine
7. Crenn P, Messing B, Cynober L. Citrulline as a biomarker of intestinal failure due to enterocyte mass reduction. Clin Nutr Edinb Scotl
8. Crenn P, Vahedi K, Lavergne-Slove A, et al. Plasma citrulline: a marker of enterocyte mass in villous atrophy-associated small bowel disease. Gastroenterology
9. Papadia C, Kelly P, Caini S, et al. Plasma citrulline as a quantitative biomarker of HIV-associated villous atrophy in a tropical enteropathy population. Clin Nutr Edinb Scot
10. Peters JHC, Beishuizen A, Keur MB, et al. Assessment of small bowel function in critical illness: potential role of citrulline metabolism. J Intensive Care Med
11. Piton G, Manzon C, Cypriani B, et al. Acute intestinal failure in critically ill patients: is plasma citrulline the right marker? Intensive Care Med
12. Piton G, Manzon C, Monnet E, et al. Plasma citrulline kinetics and prognostic value in critically ill patients. Intensive Care Med
13. Jordan JA, Durso MB. Real-time polymerase chain reaction for detecting bacterial DNA directly from blood of neonates being evaluated for sepsis. J Mol Diagn
14. Balzan S, De Almeida Quadros C, De Cleva R, et al. Bacterial translocation: overview of mechanisms and clinical impact. J Gastroenterol Hepatol
15. Wellinghausen N, Siegel D, Winter J, et al. Rapid diagnosis of candidaemia by real-time PCR detection of Candida DNA in blood samples. J Med Microbiol
2009; 58 (Pt 8):1106–1111.
16. Hansberry DR, Shah K, Agarwal P, et al. Fecal myeloperoxidase as a biomarker for inflammatory bowel disease. Cureus
17. Masoodi I, Kochhar R, Dutta U, et al. Evaluation of fecal myeloperoxidase as a biomarker of disease activity and severity in ulcerative colitis. Dig Dis Sci
18. Masoodi I, Kochhar R, Dutta U, et al. Fecal lactoferrin, myeloperoxidase and serum C-reactive are effective biomarkers in the assessment of disease activity and severity in patients with idiopathic ulcerative colitis. J Gastroenterol Hepatol
19. Kosek M, Haque R, Lima A, et al. Fecal markers of intestinal inflammation and permeability associated with the subsequent acquisition of linear growth deficits in infants. Am J Trop Med Hyg
20. Husain N, Tokoro K, Popov JM, et al. Neopterin
concentration as an index of disease activity in Crohn's disease and ulcerative colitis. J Clin Gastroenterol
21. Nancey S, Boschetti G, Moussata D, et al. Neopterin
is a novel reliable fecal marker as accurate as calprotectin for predicting endoscopic disease activity in patients with inflammatory bowel diseases. Inflamm Bowel Dis
23. Demacker PNM, Beijers AM, van Daal H, et al. Plasma citrulline measurement using UPLC tandem mass-spectrometry to determine small intestinal enterocyte pathology. J Chromatogr B Analyt Technol Biomed Life Sci
24. Brewster D, Manary M, Menzies I, et al. Intestinal permeability in kwashiorkor. Arch Dis Child
25. Semba RD, Shardell M, Trehan I, et al. Metabolic alterations in children with environmental enteric dysfunction. Sci Rep
26. Kosek MN, Mduma E, Kosek PS, et al. Plasma tryptophan and the kynurenine-tryptophan ratio are associated with the acquisition of statural growth deficits and oral vaccine underperformance in populations with environmental enteropathy. Am J Trop Med Hyg
27. Schneider RE, Viteri FE. Morphological aspects of the duodenojejunal mucosa in protein-calorie malnourished children and during recovery. Am J Clin Nutr
28. Noordally SO, Sohawon S, Semlali H, et al. Is there a correlation between circulating levels of citrulline and intestinal dysfunction in the critically ill? Nutr Clin Pract
29. Golden MH. The development of concepts of malnutrition. J Nutr
30. Baertl JM, Placko RP, Graham GG. Serum proteins and plasma free amino acids in severe malnutrition. Am J Clin Nutr
31. McCormick BJJ, Lee GO, Seidman JC, et al. Dynamics and trends in fecal biomarkers of gut function in children from 1-24 months in the MAL-ED study. Am J Trop Med Hyg
32. Faber F, Bäumler AJ. The impact of intestinal inflammation on the nutritional environment of the gut microbiota. Immunol Lett
2014; 162 (2 Pt A):48–53.
33. Kristensen KHS, Wiese M, Rytter MJH, et al. Gut microbiota in children hospitalized with oedematous and non-oedematous severe acute malnutrition in Uganda. PLoS Negl Tros Dis
34. Rytter MJH, Kolte L, Briend A, et al. The immune system in children with malnutrition–a systematic review. PLOS ONE