Efficacy of a Green Banana–Mixed Diet in the Management of Persistent Diarrhea: Protocol for an Open-Labeled, Randomized Controlled Trial (2024)

Reviewed by Christel Weiß and Khairun Nain Bin Nor Aripin

Ethics and Research Approval

This study has received approval from the Institutional Review Board (IRB) at the icddr,b (approval No. 17075, version 3.0, dated October 22, 2017). Final data will be publicly disseminated regardless of the study results. A report containing the study results will be submitted for publication in an appropriate journal after completion of data collection and analysis. The study protocol follows the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) guidance for protocol reporting [30] (see Tables 1 and ​and22).

All subjects will need to give written informed consent in accordance with the Declaration of Helsinki. The privacy, anonymity, and confidentiality of data and information identifying study participants will be strictly maintained. All medical information, description of treatment, and results from laboratory tests will be confidential and kept under lock and key; only the research staff will have access to this information. A quality assurance audit and inspection of this study may be conducted by the Ethical Committees of the IRB and will be independent of the study investigators and the sponsor; the quality assurance auditor will have access to all medical records, the investigator’s study-related files and correspondence, and the informed consent documentation relevant to this clinical study. The occurrence of any serious adverse event (eg, death) will be reported to the IRB within 24 hours of the event and their recommendations will be followed.

The study recruited patients and placed them into three different groups. An addendum to the protocol (version 4.0, dated May 5, 2018) has been approved by the IRB and includes a plan to collect stool and blood samples for microbiota and metabolomics analysis, respectively. The funder has not and will not have influence at any stage of the research, from study design to publication.

Traditionally, green banana is used as an antidiarrheal agent in the community. It is also used as a vegetable in Bangladesh, India, and other countries in Africa. Therefore, the IRB from the icddr,b did not direct us to form a data-monitoring committee.

Study Location

The study is being conducted at the Dhaka Hospital of the icddr,b, Dhaka, Bangladesh. This hospital provides care and treatment to over 166,624 patients annually of all ages and of both genders. Patients usually come with diarrheal illnesses and/or other associated problems, such as pneumonia, malnutrition, sepsis, and electrolyte abnormalities. In 2018, about 98,308 children under the age of 5 years were admitted. The majority of care seekers were from poor socioeconomic backgrounds and lived in urban and periurban Dhaka. Care is provided by a professional team, including junior and consultant physicians, nurses, counselors, and dietary workers in a multidisciplinary approach. There are different wards to treat patients with respiratory problems, diarrheal diseases, and malnutrition. The laboratory possesses well-equipped facilities capable of performing most of the clinical tests proposed in this study. For critically ill patients, there is an intensive care unit facility present within the hospital, equipped with necessary life support measures, including mechanical ventilators and syringe pumps for vasopressor support.

Study Population

This study includes patients who meet the criteria below.

Study Design

This is an open-labeled, randomized controlled clinical trial with three treatment arms (see Figure 1 and Table 3). Children 6-36 months of age admitted to the Dhaka Hospital of the icddr,b with PD or who developed PD during their treatment period and failed to respond with milk suji—a low-lactose formula made from milk powder and rice powder—have been screened and enrolled in this study. The participant enrollment period lasted 18 months.

Randomization

The permuted block randomization technique was followed to select treatment arms for each child. The randomization procedure was planned and set up by a scientist from the icddr,b who was not involved in the data collection. The randomization list containing the subject IDs and the corresponding group allocation remained concealed. IDs were chronologically assigned to each new study participant. After randomization, opaque envelopes containing the names of the allocated diet groups were opened.

Collection of Baseline Information

All children with PD, either at admission or developed at some point during their treatment period, within the defined age group were screened for study eligibility criteria by the study physician. Parents or attending caregivers of those eligible children, depending on the inclusion and the exclusion criteria, were invited to provide their consent for enrollment of their children in the study. Parents and caregivers signed a written informed consent form (see Multimedia Appendix 1) and were provided with information about the study and its interventions, possible benefits and risks, and voluntary nature of participation, including the right to withdraw children at any time after the initial consent without providing any reason; after this, children were enrolled by the study physician. One copy of the signed consent document was given to the caregiver of each participant and another copy was kept for the study documentation. A pretested case record form was used to collect relevant medical history information, including nature and duration of illness and medication for current illness. The form was also used to collect information on sociodemographic characteristics, such as age, sex, religion, gestational age, parental age, parental education, parents’ occupations, drinking water source and sanitation, fuel use and smoking history, monthly family income, number of siblings, and number of rooms in the home. Information was also collected about each child’s feeding practice, such as their history of breastfeeding, formula feedings, or other complementary feedings, as well as immunization status and each child’s past history of pneumonia and diarrhea.

Data on clinical characteristics of participants was also collected. Clinical examination measurements recorded by the study physician included pulse and respiratory rate, axillary temperature, anthropometric measurements (ie, height, weight, mid-upper-arm circumference, weight-for-age Z-score, weight-for-length Z-score, and weight-for-height Z-score), chest auscultation, oxygen saturation, presence or absence of chest-wall indrawing, cyanosis, and mental status (ie, normal, irritable, or lethargic). Weight was measured using an electronic weighing scale with a precision of 0.1 kg; height/length was measured using a locally made length board with a precision of 0.5 cm by a trained and experienced nurse from the Dhaka Hospital of the icddr,b. Fever was defined when the axillary temperature was 38°C or greater. Respiratory rate was counted for a full 60 seconds by exposing the trunk when the child was awake and calm; the presence of lower-chest-wall indrawing was noted at the same time. Frequency and consistency of stool were monitored by either the study physician or a health worker every 8 hours up to the resolution of the diarrhea.

At discharge, caregivers were asked to come back with the children for a minimum of two, weekly follow-up visits.

Laboratory Tests

All laboratory tests were carried out according to the management outline of PD based on the previous studies [31-34], which include the following tests:

1. Stool for routine microscopic examination and culture for Vibrio, Salmonella, Shigella, and Campylobacter jejuni from rectal swab culture.

2. Total and differential blood count.

3. Chest x-ray of anteroposterior view for management of pneumonia.

4. Serum electrolyte and creatinine if there is any clinical evidence of electrolyte imbalance or renal insufficiency.

5. Blood culture for suspected septicemia, typhoid fever, prolonged febrile illness, or hospital-acquired infection.

6. A rapid diagnostic test of stool by enzyme-linked immunosorbent assay (ELISA) for the diagnosis of Cryptosporidium spp and Giardia in selected cases, where the response is delayed or there is strong clinical suspicion.

With the aim to evaluate gut microbiota and proteomics in PD, additional investigations are planned as follows:

1. For gut microbiota analysis, fecal samples (2 g each) were collected from every child over the course of the study as follows: (1) on enrollment day, (2) at any time a diet was changed, and (3) at the time of discharge. Different types of microbiota of diverse groups (eg, Bacteroides, Prevotella, and Ruminococcus) will be tested by 16S rRNA sequencing.

2. For the TaqMan assay to detect other enteropathogens causing PD, a fecal sample (2 g) on screening day was collected.

For proteomic and metabolomic assays, two plasma samples (150 µL each) were collected from 50 children over the course of the study as follows: (1) on enrollment day and (2) at the time of discharge from the study; samples are to be analyzed and tested by a collaborative institute (to be decided).

Clinical Management

Children with PD were admitted to the long-stay unit of the Dhaka Hospital of the icddr,b. Initial routine investigations were completed to identify the etiology of diarrhea by performing stool routine microscopic examinations and rectal swabs; if stool routine examinations were suggestive of invasive diarrhea, an appropriate antibiotic was provided according to hospital protocol. During this period, milk suji, a low-lactose milk and rice flour-based diet containing ~67 kcal and 1.3 g protein per 100 mL, was given as a routine diet. If the PD resolved with milk suji, the child was discharged with health advice. On day 4 (ie, 3 days after milk suji was given), if diarrhea did not resolve, the child was enrolled in the study and randomization was performed. The child received one of the three diets: full-strength rice suji containing green banana, full-strength rice suji alone, or three-quarter-strength rice suji alone. Children on all three diets were followed for 7 days. If there was deterioration of diarrhea (ie, either increased frequency or watery consistency) for 3 days or if the condition remained static for up to 7 days, the child’s status was declared as treatment failure. Children whose treatment failed received dietary intervention as per the standard management of diarrhea at the Dhaka Hospital of the icddr,b (see Figure 1).

Volume of Diet

In this age group, we usually provide oral feeding equivalent to 60-85 mL/kg/day (12 feeds/24 hours). If breast milk was insufficient or the child was formula fed, the child received 120 mL/kg/day (12 feeds/24 hours). For a severely malnourished child who often did not get sufficient breast milk, the diet volume was 120 mL/kg/day (12 feeds/24 hours). Later, if a child demanded more and diarrhea had not worsened, the amount they were fed was increased to 144 mL/kg/day. The consultant physician made the final decision about the dietary volume, depending on the clinical condition of each patient. The volume offered and actual intake were properly recorded.

Follow-Up After Discharge

Children who were fed full-strength rice suji, with or without green banana, or three-quarter-strength rice suji were discharged and were required to follow up for 2 weeks. At the end of 14 days, they returned to the hospital and if they remained diarrhea free, the diet was switched back to milk suji and caregivers were advised to introduce other family diet items gradually. Children with severe acute malnutrition and severe pneumonia received treatment according to the hospital’s standard management protocol [35,36] and WHO guidelines [32], respectively. Last but not the least, it is important to mention that caregivers who were the mothers of the children were encouraged to continue breastfeeding along with providing their children with a specific diet. The food in the study was prepared and provided by health workers under the close supervision of a qualified dietician; the diets were formulated based on locally available, culturally acceptable, affordable foods, quite similar to the WHO’s recommendations [3,14,16].

Outcome Measures

Sample Size

Rabbani et al [20] conducted a study where they enrolled 5-12-month-old children; they found that their recovery rates from PD by day 4 on the green banana diet and rice suji was 78% and 23%, respectively. However, studies by Islam et al [13] and Mahfuz et al [37] found that a higher percentage of children recovered from PD after being given rice suji. There has been no study conducted where diets of rice suji with or without green banana were given to children 6-36 months of age. Considering these facts, we assumed that in 6-36-month-old children, the rate of recovery from diarrhea by day 5 in either of the intervention diet groups (ie, full-strength rice suji with or without green banana) would be 90% and that the rate of recovery in the control diet group (ie, three-quarter-strength rice suji) would be 65%.

With 80% power and a 5% type I error, and considering three treatment arms, we needed 40 children in each arm. If we enrolled 45 children in each arm, that would accommodate up to an 11% rate of attrition. Therefore, the total target sample size was 135 children (ie, 45 children × 3 arms).

Statistical Analysis

Data has been entered into a personal computer and will be analyzed using SPSS Statistics for Windows, version 20.0 (IBM Corp). Statistical analyses will include descriptive as well as analytical methods. We will compare different characteristics (eg, age, gender, diarrhea and its duration, stool consistency, presence or absence of blood, fast breathing, lower-chest-wall indrawing, and fever). We will also evaluate the PD outcome as improved, death in hospital, relapse, or death during follow-up. Categorical variables will be compared using the chi-square test. When the variables of interest are continuous and parametric, the statistical significance of group mean comparisons will be evaluated by analysis of variance (ANOVA). When the main outcome measures are continuous nonparametric variables, the statistical significance of differences will be determined using the Kruskal-Wallis test. The post hoc test will be done accordingly. A probability of less than .05 will be considered statistically significant. Strength of association will be determined by calculating relative risk and 95% CI. Our primary analysis is an intention-to-treat analysis comparing the groups. Survival analysis will also be carried out. Finally, regression analyses will be performed to reach more definitive conclusions.

Multimedia Appendix 1

Participant consent form.

Multimedia Appendix 2

Previous peer review reports from icddr,b.

Conflicts of Interest: None declared.

1. Walker CLF, Rudan I, Liu L, Nair H, Theodoratou E, Bhutta ZA, O'Brien KL, Campbell H, Black RE. Global burden of childhood pneumonia and diarrhoea. Lancet. 2013 Apr 20;381(9875):1405–1416. doi:10.1016/S0140-6736(13)60222-6. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

4. Rahman AE, Moinuddin M, Molla M, Worku A, Hurt L, Kirkwood B, Mohan SB, Mazumder S, Bhutta Z, Raza F, Mrema S, Masanja H, Kadobera D, Waiswa P, Bahl R, Zangenberg M, Muhe L, Persistent Diarrhoea Research Group Childhood diarrhoeal deaths in seven low- and middle-income countries. Bull World Health Organ. 2014 Sep 01;92(9):664–671. doi:10.2471/BLT.13.134809. http://europepmc.org/abstract/MED/25378757. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

5. Mbori-Ngacha DA, Otieno JA, Njeru EK, Onyango FE. Prevalence of persistent diarrhoea in children aged 3-36 months at the Kenyatta National Hospital, Nairobi, Kenya. East Afr Med J. 1995 Nov;72(11):711–714. [PubMed] [Google Scholar]

6. Sodeinde O, Adeyemo A, Gbadegesin R, Ademowo O. Persistent diarrhoea in Nigerian children aged less than five years: A hospital-based study. J Diarrhoeal Dis Res. 1997 Sep;15(3):155–160. [PubMed] [Google Scholar]

8. Mathai J, Raju B, Bavdekar A, Pediatric Gastroenterology Chapter‚ Indian Academy of Pediatrics Chronic and persistent diarrhea in infants and young children: Status statement. Indian Pediatr. 2011 Jan;48(1):37–42. doi:10.1007/s13312-011-0018-9. [PubMed] [CrossRef] [Google Scholar]

9. Das S, Faruque A, Chisti M, Malek M, Salam M, Sack D. Changing trend of persistent diarrhoea in young children over two decades: Observations from a large diarrhoeal disease hospital in Bangladesh. Acta Paediatr. 2012 Oct;101(10):e452–e457. doi:10.1111/j.1651-2227.2012.02761.x. [PubMed] [CrossRef] [Google Scholar]

10. Mahalanabis D, Alam AN, Rahman N, Hasnat A. Prognostic indicators and risk factors for increased duration of acute diarrhoea and for persistent diarrhoea in children. Int J Epidemiol. 1991 Dec;20(4):1064–1072. doi:10.1093/ije/20.4.1064. [PubMed] [CrossRef] [Google Scholar]

12. Victora C, Huttly S, Fuchs S, Barros FC, Garenne M, Leroy O, Fontaine O, Beau JP, Fauveau V, Chowdhury HR. International differences in clinical patterns of diarrhoeal deaths: A comparison of children from Brazil, Senegal, Bangladesh, and India. J Diarrhoeal Dis Res. 1993 Mar;11(1):25–29. [PubMed] [Google Scholar]

13. Islam SB, Ahmed T, Mahfuz M, Mostafa I, Alam MA, Saqeeb KN, Sarker SA, Chisti MJ, Alam NH. The management of persistent diarrhoea at Dhaka Hospital of the International Centre for Diarrhoeal Disease and Research: A clinical chart review. Paediatr Int Child Health. 2018 May;38(2):87–96. doi:10.1080/20469047.2017.1315911. [PubMed] [CrossRef] [Google Scholar]

15. Akbar MS, Roy SK, Banu N. Persistent diarrhoea: Management in algorithmic approach using a low-cost rice-based diet in severely malnourished Bangladeshi children. J Trop Pediatr. 1993 Dec;39(6):332–337. doi:10.1093/tropej/39.6.332. [PubMed] [CrossRef] [Google Scholar]

17. Roy S, Akramuzzaman S, Haider R, Majid N, Khatun M, Akbar M, Alam A. Persistent diarrhoea: Factors affecting absorption and clinical prognosis during management with a rice-based diet. Acta Paediatr Suppl. 1992 Sep;381:139–143. doi:10.1111/j.1651-2227.1992.tb12388.x. [PubMed] [CrossRef] [Google Scholar]

18. Roy SK, Alam AN, Majid N, Khan AM, Hamadani J, Shome GP. Persistent diarrhoea: A preliminary report on clinical features and dietary therapy in Bangladeshi children. J Trop Pediatr. 1989 Apr;35(2):55–59. doi:10.1093/tropej/35.2.55. [PubMed] [CrossRef] [Google Scholar]

19. Rabbani GH, Teka T, Zaman B, Majid N, Khatun M, Fuchs GJ. Clinical studies in persistent diarrhea: Dietary management with green banana or pectin in Bangladeshi children. Gastroenterology. 2001 Sep;121(3):554–560. doi:10.1053/gast.2001.27178. [PubMed] [CrossRef] [Google Scholar]

20. Rabbani GH, Teka T, Saha SK, Zaman B, Majid N, Khatun M, Wahed MA, Fuchs GJ. Green banana and pectin improve small intestinal permeability and reduce fluid loss in Bangladeshi children with persistent diarrhea. Dig Dis Sci. 2004 Mar;49(3):475–484. doi:10.1023/b:ddas.0000020507.25910.cf. [PubMed] [CrossRef] [Google Scholar]

22. Basu S, Chatterjee M, Ganguly S, Chandra PK. Effect of Lactobacillus rhamnosus GG in persistent diarrhea in Indian children: A randomized controlled trial. J Clin Gastroenterol. 2007 Sep;41(8):756–760. doi:10.1097/01.mcg.0000248009.47526.ea. [PubMed] [CrossRef] [Google Scholar]

23. Faisant N, Buléon A, Colonna P, Molis C, Lartigue S, Galmiche J, Champ M. Digestion of raw banana starch in the small intestine of healthy humans: Structural features of resistant starch. Br J Nutr. 1995 Jan;73(1):111–123. [PubMed] [Google Scholar]

24. Englyst H, Cummings J. Digestion of the carbohydrates of banana (Musa paradisiaca sapientum) in the human small intestine. Am J Clin Nutr. 1986 Jul;44(1):42–50. doi:10.1093/ajcn/44.1.42. [PubMed] [CrossRef] [Google Scholar]

25. Rabbani G, Albert M, Rahman H, Chowdhury A. Short-chain fatty acids inhibit fluid and electrolyte loss induced by cholera toxin in proximal colon of rabbit in vivo. Dig Dis Sci. 1999 Aug;44(8):1547–1553. doi:10.1023/a:1026650624193. [PubMed] [CrossRef] [Google Scholar]

26. Binder HJ. Role of colonic short-chain fatty acid transport in diarrhea. Annu Rev Physiol. 2010;72:297–313. doi:10.1146/annurev-physiol-021909-135817. [PubMed] [CrossRef] [Google Scholar]

27. Binder HJ, Mehta P. Short-chain fatty acids stimulate active sodium and chloride absorption in vitro in the rat distal colon. Gastroenterology. 1989 Apr;96(4):989–996. doi:10.1016/0016-5085(89)91614-4. [PubMed] [CrossRef] [Google Scholar]

29. Sarker SA, Ahmed T, Brüssow H. Persistent diarrhea: A persistent infection with enteropathogens or a gut commensal dysbiosis? Environ Microbiol. 2017 Oct;19(10):3789–3801. doi:10.1111/1462-2920.13873. [PubMed] [CrossRef] [Google Scholar]

30. Chan A, Tetzlaff J, Altman D, Laupacis A, Gøtzsche PC, Krle A-Jerić K, Hrobjartsson A, Mann H, Dickersin K, Berlin JA, Dore CJ, Parulekar WR, Summerskill WS, Groves T, Schulz KF, Sox HC, Rockhold FW, Rennie D, Moher D. SPIRIT 2013 Statement: Defining standard protocol items for clinical trials. Rev Panam Salud Publica. 2015 Dec;38(6):506–514. http://europepmc.org/abstract/MED/27440100. [PMC free article] [PubMed] [Google Scholar]

31. Bhutta ZA, Ghishan F, Lindley K, Memon IA, Mittal S, Rhoads JM, Commonwealth Association of Paediatric Gastroenterology and Nutrition Persistent and chronic diarrhea and malabsorption: Working Group report of the second World Congress of Pediatric Gastroenterology, Hepatology, and Nutrition. J Pediatr Gastroenterol Nutr. 2004 Jun;39 Suppl 2:S711–S716. doi:10.1097/00005176-200406002-00019. [PubMed] [CrossRef] [Google Scholar]

33. Bardhan PK, Albert MJ, Alam NH, Faruque SM, Neogi PK, Mahalanabis D. Small bowel and fecal microbiology in children suffering from persistent diarrhea in Bangladesh. J Pediatr Gastroenterol Nutr. 1998 Jan;26(1):9–15. doi:10.1097/00005176-199801000-00002. [PubMed] [CrossRef] [Google Scholar]

35. Ahmed T, Ali M, Ullah MM, Choudhury IA, Haque ME, Salam MA, Rabbani GH, Suskind RM, Fuchs GJ. Mortality in severely malnourished children with diarrhoea and use of a standardised management protocol. Lancet. 1999 Jun 05;353(9168):1919–1922. doi:10.1016/S0140-6736(98)07499-6. [PubMed] [CrossRef] [Google Scholar]