Chromatographic immunoassays for Helicobacter pylori detection – are they reliable in Mali, West Africa?

Chromatographic immunoassays for Helicobacter pylori detection – are they reliable in Mali, West Africa?

 

Ingvild Austarheim^1,&, Kari Tvete Inngjerdingen¹, Berit Smestad Paulsen¹, Adiaratou Togola², Chiaka Diakité², Drissa Diallo^1,2

 

¹School of Pharmacy, University of Oslo, P.O. Box 1068 Blindern, 0316 Oslo, Norway, ²Department of Traditional Medicine, BP 1746, Bamako, Mali

 

 

^&Corresponding author
Ingvild Austarheim, School of Pharmacy, P.O.Box 1068 Blindern, 0316 Oslo, Norway

 

 

 

 

Gastritis and other gastrointestinal (GI) disorders are important causes of morbidity in the African population [1]. In Mali, the prevalence of gastric ulcer in the population is reported to be 4.2% for men and 2.4% for women, and is probably higher for gastritis and non-ulcer dyspepsia [2,3]. Medicinal plants are popular remedies for treating gastrointestinal ailments in Mali, as approximately 75% of the population mainly use plant based medicines for their primary health care [2,4,5]. It is important to increase the knowledge behind the aetiology of gastritis, gastric ulcer and GI-related problems in Mali for future research on Malian traditional medicines used for these ailments, and also in order to give patients the correct treatment.The presence of Helicobacter pylori in the population is one important factor needing assessment, as H. Pylori is a known factor contributing to dysfunction of the GI.

Individuals may be asymptomatic carriers of the disease, but the presence of H. Pylori is highly correlated with underlying ailments causing dyspepsia. Infections with H. Pylori are found worldwide in individuals of all ages, but are commonly acquired at an earlier age in developing countries [6]. A previous study conducted in Mali on patients with gastric ulcer reported an H. Pylori prevalence of 95% [7]. The prevalence of H. Pylori in dyspeptic patients in other West African countries, diagnosed by biopsy and urease test, is reported to be 97% in Ghana [8], 91.3% in Ivory Coast [9], and 72-91% in Nigeria [10,11]. In Senegal a prevalence of 100% was found in patients with ulcers or gastritis by urea breath test and histological findings [12].

Rapid stool- or serology-based chromatographic immunoassays were chosen in this study, as we wanted to diagnose patients in Mali with a simple, non-invasive technique.

 

 

Volunteers with confirmed gastric ulcer

Twenty-nine volunteers diagnosed with gastric ulcer, 18 (62%) male and 11 (38%) female, confirmed by gastroscopy at Cabinet Medical "Les étoiles" at Gabriel Touré hospital, Bamako, were tested by the chromatographic immune based assay Clearview^®H. Pylori (Alere Inc., Waltham, MA). Nine different ethnic groups were included; Bambara (31%), Peulh (28%), Sénoufo, Malinké, Somono, Dogon, Dafing, Sonrhaï and Minianka. The average age was 41 years (age 27 - 65 years). H. Pylori antibodies (IgG) present in fresh whole blood were analysed according to the manufacturer?s instructions. Briefly, a thin lancet was used to puncture the skin of the finger and 50 µl whole blood and one drop of buffer was dispensed into the round window of the test device. The assay was carried out at room temperature (30°C), and was read after 10 minutes. A pink-red line, even very weak, was considered as a positive result. All volunteers gave informed consent, confirmed by a signature.

Randomly selected volunteers

The chromatographic immune based assay Clearview^®H. pylori was used to detect H. pylori antibodies (IgG) present in fresh whole blood of 59 volunteers. The volunteers were randomly recruited from three different areas of Bamako, Mali; Hippodrome, Daoudabougou and Sotuba. The group included 31 (53%) female and 28 (47%) male with an average age of 36 years (age 18 - 65 years). Fourteen different ethnic groups were involved; Bambara (24%), Malinké (14%), Bozo, Dafing, Haoussa, Kakolo, Peulh, Sarakolé, Sénoufo, Sonrhaï, Tamasheq, Fofana, Sangare and Minianka. Whole blood was analysed as described under subsection 2.1. All volunteers gave informed consent, confirmed by a signature or a fingerprint.

Volunteers with diverse gastric ailments

The chromatographic immune based assay ImmunoCard STAT!^®HpSA^® Stool antigen test (Meridian Bioscience Inc., Cincinnati, OH) was used to detect H. pylori antigens present in the stool of 65 volunteers. The stool samples were given at the INRSP (Institut National de Recherche en Santé Publique, Bamako) due to ailments like infections (33%), dyspepsia (16%) and other stomach related ailments. The group included 33 (52%) female and 31 (48%) male, and with an average age of 33 years (age 2 - 70 years). Thirteen different ethnic groups were included; Sarakolé (23%), Peulh (20%), Bambara (19%), Sénoufo, Dogon, Malinké, Kasonké, Kakolo, Maure, Minianka, Diawando, Sonhraï and Somono. All stool samples had a pasty consistency and they were analysed at room temperature (25°C) two hours after stool sampling, according to the manufacturer?s guidelines. Briefly, four drops of diluted stool were dispensed into the round window of the test device and the result was read after five minutes of incubation at room temperature. The appearance of a pink-red band in the reading window, even very weak, was considered as a positive result. All volunteers gave an informed consent for bacteriological research on stool samples.

Ethical approval

This work was a research project integrated in a diagnostic process at INRSP and in research program on traditional anti-gastric ulcer medicinal plants. For this reason, the project on its own was not presented to the ethic Committee of INRSP. Prior consent was obtained from all participants before taking or using samples

 

 

Volunteers with confirmed gastric ulcer

H. pylori antibodies were detected in 21% of the 29 volunteers diagnosed with gastric ulcer by gastroscopy (Table 1). These individuals were tested directly after endoscopy, and had not been through H. Pylori eradicating regime. No association with a positive detection of H. Pylori and gender, age or ethnic group was found (Additional material A).

Randomly selected volunteers

H. pylori antibodies were detected in 44% of the randomly selected volunteers (Table 1). No association between H. Pylori positive individuals, gender, age or ethnic group was found. Nine (15.2%) individuals reported that they had previously undergone gastroscopy; out of these two were H. Pylori positive in the present study. Thirty-six (61%) of the volunteers reported that they from time to time experienced dyspepsia, mostly due to Gastroesophageal Reflux Disease (GARD), and of these individuals 36% were H. pylori positive. Among the volunteers who did not experience GI symptoms, 54% were H. pylori positive. Four of the volunteers knew that they had recently been taking antibiotics, and of these three were H. Pylori positive. See Additional material B for additional data on gender, age, ethnicity and GI related problems.

Volunteers with diverse gastric ailments

H. Pylori antigens were present in 14% of the volunteers suffering from diverse gastric ailments (Table 1). Of the nine individuals that were H. Pylori positive, five were tested for gastrointestinal infections problems. None of the dyspeptic individuals tested positive (Additional material C).

 

 

According to the manufacturer, Clearview^®H. pylori is regarded as a sensitive test ( 2), and a negative result should therefore indicate that the person truly is negative. In the present study we found that only 21% of the individuals with confirmed gastric ulcer tested positive for H. pylori by using Clearview^®H. pylori . This number is probably too low to be realistic as it does not correlate with a previous investigation carried out in Mali in year 2000. Biopsies taken from 121 patients diagnosed with gastric ulcer by endoscopy confirmed that H. Pylori were present in 95% of the cases [7].The population studied came from the same area (Bamako) and the gastric ulcer was diagnosed by endoscopy in the same manner as in the present study. The culture and habits have not changed much in Mali the last decade, which means that the populations studied should be comparable. It would therefore be expected that most of the individuals diagnosed with gastric ulcer in the current study should be H. pylori positive.

A low (41%) prevalence was also shown by Clearview^®H. pylori among the randomly selected volunteers. The H. pylori sero-prevalence of healthy individuals in nearby countries is reported to be much higher, 80 - 85% in Nigeria [13] and 75.4% in Benin [14]. In addition, random studies in Africa have shown a seroprevalence of 80% [1]. Factors that seem to play a role in the infection rate are low-quality drinking water, lack of basic hygiene, habit of eating from the same plate as well as poor diet and overcrowding [11,15]. These factors are similar between Mali and the surrounding countries, making a comparable prevalence highly likely. To our comprehension, the most obvious reason to explain the low detection of H. Pylori is that Clearview^®H. pylori has a low sensitivity in the population studied in Mali. Sensitivity is not an intrinsic property of an assay [16], and it is therefore possible that the sensitivity experienced in the population studied here, is not the same as for the population studied by the manufacturer.

A low prevalence of H. pylori (14%) was also found by the ImmunoCard STAT!^®HpSA^® stool test. A good practice in diagnosing H. pylori with antigen tests is that antibiotics should be stopped at least four weeks and anti-secretory therapy at least two weeks before testing, as these medications negatively affects the sensitivity [11]. In Mali malaria is endemic, and anti-malaria drugs might have a negative effect on the sensitivity of the test, as these drugs might eradicate or decrease the presence of H. pylori [17,18]. Most of the individuals in our study have a low educational level, and it was therefore difficult to know if they had been taking any interfering medications. H. pylori is, however, not easily eradicated [19], and self-medication is probably not sufficient for total removal of the bacteria, but potentially enough to lower the amount of stool antigens to make the outcome negative.

According to Meridian Bioscience, it is a possibility that the stool test does not respond to all subgroups of H. Pylori . There is a remarkable genetic diversity of H. Pylori worldwide, the number of genetic variability of H. Pylori exceeding 2 to 5-fold the variability of other bacterial pathogens. Substantial diversity has recently also been described for the important virulence factor cagA [20]. This means that ImmunoCard STAT!^®HpSA^®, which uses monoclonal antibodies for detection, might have a lower sensitivity towards antigens produced by some strains. Calvet et al. [21] reported a sensitivity of 69-74% for ImmunoCard STAT!^®HpSA^® on patients with dyspeptic symptoms. This is much lower then what reported by the manufacturer (Table 2).

A weakness in the present study might be that the serological and the stool tests were not carried out on the same individuals, thereby making it difficult to compare these methods directly. However, our results suggest that both assays have a considerably lower sensitivity in Mali than reported by the manufacturers. A follow-up study is needed in order to give any definite conclusions. It is noteworthy that the majority (61%) of the randomly selected volunteers experienced gastrointestinal symptoms. This reflects the high degree of GI related ailments experienced by the population of Mali. Natural products which reduce the density of H. pylori colonization and mucus adherence [22], and thereby lower the dyspeptic symptoms, might be highly relevant for further research as rapid recurrence and antibiotic resistance are increasing problems in countries with high a prevalence of H. pylori [23]. In addition, as Mali is ranked as one of the poorest countries in the world, the population has medicinal plants as the main option for most of primary health care. It is therefore highly important to carry out more research on traditional medicines used for these types of ailments, and also to have access to simple and non-expensive diagnostically methods reliable for the Malian population.

 

 

Rapid immunoassays forH. Pylori detection need to be properly evaluated in the respective countries where they are to be used before they can be relied upon as diagnostic tools. Interfering circumstances such as medications and strain variability should be addressed in greater detail. Due to endemic malaria, strain variability and self medication, it is a possibility that the stool based test looses some of its sensitivity. As self medication is difficult to avoid, it is probably wise to choose a different test system for H. Pylori identification in Mali.

 

 

There is no competing interests in this paper.

 

 

Ingvild Austarheim has performed all the serological testes (in Mali) for the randomly selected volunteers, including interviews. Together with Kari Inngjerdingen she has participated in discussions, and prepared the whole manuscript and finalised it. Adiaratou Togola has been involved in the testing of the samples for the stool tests. Drissa Diallo has contributed by discussions of protocol and revising the paper and by being responsible for the testing of the biological samples both at DMT and at INRSP. Chiaka Diakité has been selecting the persons to be tested for blood sample testing, and he was medically responsible for the project. Berit Smestad Paulsen was involved in discussions of the protocols and the set up of the project, also taking part in discussions for the writing of the paper. The final version of the article was approved by all authors.

 

 

The Norwegian Research Council, project number 190140/V10 and the EU project FP7 no. 266005, Multi-disciplinary University Traditional Health Initiative (MUTHI): Building Sustainable Research Capacity on Plants for Better Public Health in Africa, are acknowledged for financial support. The Norwegian Pharmaceutical Society is acknowledged for covering expenses during the first authors´ travel to Mali. Fagnan Sanogo and Fatoumata Kaou Sissoko are acknowledged for their technical assistance and all the volunteers are acknowledged for making this survey possible. Seydou Diarra and Moussa Sacko are acknowledged for letting us use the lab facilities at department of diagnosis and bacteriological research, Bamako.

 

 

Table 1: Results obtained from serological and stool based tests in Helicobacter pylori detection

Table 2: Test sensitivities of the utilized serological and stool based tests as given by the manufacturer

 

 

Additional material A - Pdf(21 Kb): Confirmed gastric ulcer by gastroscopy (serological test, Clearview^®H. pylori)

Additional material B - Pdf(21 Kb): Randomly selected volunteers (serological test, Clearview^® H. pylori)

Additional material C - Pdf(21 Kb): Volunteers seeking help for diverse gastric ailments (stool test, ImmunoCard STAT!^®HpSA^®)

 

 

1. Holcombe C. Helicobacter pylori: the African enigma. Gut. 1992; 33(4):429-43. PubMed | Google Scholar


2. Touré AK. Evaluation de l?efficacité thérapeutique d?une recette traditionnelle améliorée «le Gastrosedal» dans le traitement des gastrites. University of Bamako; Mali, M Sc dissertation. 1989. PubMed | Google Scholar


3. Maïga MY, Guindo S, Traore HA, Dembele M, Guindo A, Kalle A, Pichard E. Etude épidémiologique des affections oeso-gastro-duodénales au Mali au moyen de la fibroscopie digestive haute. Médecine d?Afrique noire. 1995; 42:68-72. PubMed | Google Scholar


4. Diallo D, Coumaré A, Koïta N. Etude préliminaire d´une plante médicinale du Mali: Vernonia kotschyana (Sch Bip). Cahier spécial de l´ INRSP. 1990; (1):52-56. PubMed | Google Scholar


5. Robison MM, Zhang X. The world medicine situation 2011, Traditional medicines: Global situation, issues and challenges (World Health Organization). http://apps.who.int/medicinedocs/documents/s18063en/s18063en.pdf. Accessed 21 May 2012. PubMed | Google Scholar


6. Wang AY, Peura DA. The prevalence and incidence of Helicobacter pylori -associated peptic ulcer disease and upper gastrointestinal bleeding throughout the world. Gastrointest Endosc Clin N Am. 2011; 21(4):613-635. PubMed | Google Scholar


7. Mourtala I. Infection à Helicobacter pylori et les pathologies oeso-gastroduodenales dans le centre d?endoscopie digestive de l?hôpital National du Point « G ». University of Bamako, Mali, M Sc dissertation. 2000. PubMed | Google Scholar


8. Kidd M, Louw JA, Marks IN. Helicobacter pylori in Africa: observations on an 'enigma within an enigma'. J Gastroenterol Hepatol. 1999; 14(9):851-85. PubMed | Google Scholar


9. Diomande MI, Flejou JF, Potet F, Dago-Akribi A et al. Chronic gastritis and Helicobacter pylori infection on the Ivory Coast: A series of 277 symptomatic patients. Gastroenterol Clin Biol. 1991; 15(10):711-716. PubMed | Google Scholar


10. Jemilohun AC, Otegbayo JA, Ola SO, Oluwasola OA, Akere A. Prevalence of Helicobacter pylori among Nigerian patients with dyspepsia in Ibadan. Pan Afr Med J. 2010; 6:18. PubMed | Google Scholar


11. Hunt RH, Xiao SD, Megraud F, Leon-Barua R, Bazzoli F et al. World Gastroenterology Organisation Global Guidelines: Helicobacter pylori in developing countries. http://www.worldgastroenterology.org/assets/downloads/en/pdf/guidelines/11_helicobacter_pylori_developing_countries_en.pdf. Accessed 24 April 2012. PubMed | Google Scholar


12. Mbengue M, Diouf ML, Dangou JM, Ka MM, Ba-Seck A et al. Frequency of Helicobacter pylori infection in symptomatic patients in Senegal. Med Trop (Mars). 1997; 57:256-258. PubMed | Google Scholar


13. Oluwasola AO, Ogunbiyi JO. Gastric cancer: aetiological, clinicopathological and management patterns in Nigeria. Niger J Med. 2003; 12(4):177-186. PubMed | Google Scholar


14. Aguemon BD, Struelens MJ, Massougbodji A, Ouendo EM. Prevalence and risk-factors for Helicobacter pylori infection in urban and rural Beninese populations. Clin Microbiol Infect. 2005; 11:611-617. PubMed | Google Scholar


15. Tanih NF, Dube C, Green E, Mkwetshana N, Clarke AM et al. An African perspective on Helicobacter pylori: prevalence of human infection, drug resistance, and alternative approaches to treatment. Ann Trop Med Parasitol. 2009; 103(3):189-204. PubMed | Google Scholar


16. Lunet N, Peleteiro B, Carrilho C, Figueiredo C, Azevedo A. Sensitivity is not an intrinsic property of a diagnostic test: empirical evidence from histological diagnosis of Helicobacter pylori infection. BMC Gastroenterol. 2009; 9:98. PubMed | Google Scholar


17. Goswami,S, Bhakuni RS, Chinniah A, Pal A, Kar SK et al. Anti Helicobacter pylori potential of artemisinin and its derivatives. Antimicrob Agents Chemother. 2012; in press. PubMed | Google Scholar


18. Cherian S, Forbes D, Sanfilippo F, Cook A, Burgner D. The epidemiology of Helicobacter pylori infection in African refugee children resettled in Australia. Med J Aust. 2008; 189(8):438-441. PubMed | Google Scholar


19. Huang JQ, Hunt RH. Review: Eradication of Helicobacter pylori: Problems and recommendations. Journal of Gastroenterology and Hepatology. 1997; 12(8):590-598. PubMed | Google Scholar


20. Varbanova M, Schulz C, Malfertheiner P. Helicobacter pylori and other gastric bacteria. Dig Dis. 2011; 29(6):562-569. PubMed | Google Scholar


21. Calvet X, Sanchez-Delgado J, Montserrat A, Lario S et al. Accuracy of Diagnostic Tests for Helicobacter pylori: A Reappraisal. Clin Infect Dis. 2009; 48(10):1385-1391. PubMed | Google Scholar


22. Wittschier N, Faller G, Hensel A. Aqueous extracts and polysaccharides from Liquorice roots (Glycyrrhiza glabra L.) inhibit adhesion of Helicobacter pylori to human gastric mucosa. J Ethnopharmacol. 2009; 125(2):218-223. PubMed | Google Scholar


23. Ahuja V, Sharma MP. High recurrence rate of Helicobacter pylori infection in developing countries. Gastroenterology. 2002; 123(2):653-654. PubMed | Google Scholar