De los directivos

Past medical and family histories

All

subjects H. pylori

p RR 95% CI

Positive Negative

n (%) n (%) n (%)

Yes 7 1 (14.3) 6 (85.7) 0.335 0.36 0.04, 3.08

No 223 70 (31.4) 153 (68.6)

Previous

admission(s)

Yes 50 15 (30.0) 35 (70.0) 0.656 0.86 0.43, 1.69

No 180 56 (31.1) 124 (68.9)

History of

infantile colic

Yes 33 13 (39.4) 20 (60.6) 0.252 1.55 0.72, 3.34

No 197 58 (29.4) 139 (70.6)

History of antacid use

Yes 18 11 (61.1) 7 (38.9) 0.004 3.98 1.47, 9.78

No 212 60 (28.3) 152 (71.7)

Duration of antacid use

1 week 9 4 (44.4) 5 (55.6) 0.335 0.23 0.03, 1.77

≥2 weeks 9 7 (77.8) 2 (22.2)

Family history of dyspepsia

Yes 41 21 (51.2) 20 (48.8) 0.002 2.92 1.46, 5.83

No 189 50 (26.5) 139 (73.5)

In this prospective, observational seroepidemiologic study of H. pylori, a total of 230 children admitted into the children emergency unit of the University of Uyo Teaching Hospital, southern Nigeria, had their serum samples tested for anti-H. pylori immunoglobulin G (IgG). Seventy-one out of the 230 subjects were seropositive, giving a seroprevalence rate of 30.9%.

This seroprevalence rate is high and suggests that H. pylori infection is significant in the paediatric age group of the study locality. This is consistent with high prevalence rates reported among children in other developing countries. In China, the seroprevalence rate of a childhood population of 7-14 years old school children was 40.9%.⁴⁰ Langat et al,¹¹⁶ reported a prevalence of 45.6% in a young childhood population of under 3 years old children in Kenya, eastern Africa. In contrast, studies done in the developed countries have reported low prevalence rates, in their childhood populations. Malaty et al,¹⁹ reported a childhood seroprevalence rate of 19% in the United States of America. GranstrÖm et al,¹² reported a seroprevalence rate of 13.6%

in a cohort study of Swedish school children. The higher prevalence rates in developing countries is thought to be a consequence of the poor socioeconomic conditions prevalent in these countries.³

The data also showed that H. pylori infection is acquired early in this study population, as 30.0% of infants less than 1 year old, were seropositive. The highest age-specific prevalence of 37.5% was seen in the 5.0-9.9 years age group, while the lowest

prevalence of 25% was seen in the ≥10.0 years age group. This trend contrasts with the reports of increasing prevalence rate with increasing age from some other studies.^10,19,40

There was no association between H. pylori seropositivity with the height, weight, body mass index and temperature of the subjects. Reports in the literature have not been consistent on the association of growth parameters and H. pylori infection in children. Oderda et al ⁹⁰ did not find H. pylori infection to be a risk factor for short stature in children after controlling for the socio-economic status of the family, but Choe et al ⁹¹ reported that the mean height of H. pylori -infected school children was significantly lower in comparison to that of H. pylori-negative children.

Though the prevalence rate obtained in this study is high, when compared to the age specific prevalence rate of 82% reported in northern Nigerian children aged 5 – 10 years in 1992,¹⁰ it is much lower. This lower prevalence rate may represent an actual decrease in the prevalence of H. pylori infection. Studies have shown a trend of decreasing prevalence over the past one to two decades in both adult and paediatric populations. A report of studies from an adult population in Greece,¹¹⁷ noted a decreasing seroprevalence rate of about 15 – 24% over a 10 year period (1987 - 1997) in each of the age groups followed up in the study. A similar trend was also reported in a paediatric seroepidemiologic survey in Guangzhou, southern China, between 1993 and 2003.¹¹⁸ The overall age standardization seroprevalence rate was reported to have

dropped from 30.8% in 1993 to 19.4% in 2003, among children aged 1-5 years. Thus, a decreasing seroprevalence level of 13.2% was recorded. This trend of decreasing seroprevalence of H. pylori infection has been attributed to an improvement in the general living conditions of these countries. The comparatively lower prevalence rate obtained in this study in comparison with what was found in Maiduguri about 17 years earlier may be due to a similar reason of improved living conditions over these past years. Another study in the country, more recent than the Maiduguri study, also reported a lower prevalence rate. Ugwuja and Ugwu,¹¹⁹ in 2007, found a seroprevalence of 11% among patients aged ≤20 years in Abakaliki, south-eastern Nigeria. The differences in prevalence rates could also be due to differences in the study populations. The Maiduguri study was community-based while those of Abakaliki and Uyo were hospital-based studies.

There was no gender differences in the seroprevalence rates of this study and this observation has been reported by many other authors.3,6,9,10,12,14 A contrasting report was however reported by Xu et al, ³⁹ where a significantly higher prevalence rate was reported in the male subjects. The subject selection for that study was however uneven in relation to gender, rural/urban residence and type of school attended (public/private). The study subjects were children selected from one rural private school against two schools, each from a private and public school in an urban setting. The number of subjects from each setting was not stated and the study reported a significantly higher seroprevalence rate among rural compared to urban subjects. Again

the gender distribution in relation to the age of the subjects was not stated whereas seropositivity was reported to increase with increasing age. There is a possibility therefore, that if more of the male subjects were older and/or from the rural setting then the higher seroprevalence among males could be a biased report.

The seroprevalence rate observed in this study, showed a trend of an inverse significant association with the socioeconomic status of the parents when considering social classes II to V. The study population is in a developing country and within the study population; a significantly higher seroprevalence was noted with the lowest social class.

The social class was derived from the occupation and educational attainment of parents. These observations lend support to the fact that low socioeconomic status impacts positively on the prevalence of this infection.

Considering educational levels, the EUROGAST study of 1993,¹²⁰ showed that H. pylori seropositivity had a significant inverse trend with the standard of education (p<00001).

The prevalence increased from 34% in subjects with a higher education to 47% in subjects with secondary education and 62% in subjects with only primary school education.

Low socioeconomic background and its natural consequences of overcrowding, poor hygiene and absent or insufficient sanitation have been reported to predispose to acquisition of H. pylori infection.^121-123 In a prospective population-based study of

1,545 Czech children aged 0 to 15 years, Sýkora,¹²² reported that the prevalence of H.

pylori infection was higher in socioeconomically disadvantaged institutionalised children compared to healthy children living in family units. Socioeconomic variation is the major reason advanced to explain the relatively high prevalence rates in the developing compared to developed countries of the world.^3,12,42 A contrasting finding was however reported from a seroepidemiologic study of H. pylori infection among school children and teachers in Taiwan,¹²⁴ where there was no association with socioeconomic status. This observation could have been due to subjects having a narrow gap in their socioeconomic status. Taiwan is a developed country and the government provides economic support for unemployed persons and families with children.¹²⁴ The study determined the socioeconomic status of teachers and the parents of the children from employment and educational levels. Therefore, the socioeconomic differences may not be as marked as what obtains in developing countries where the unemployed have a very low socioeconomic status.

The seroprevalence of H. pylori was significantly associated with increased household population. Households with 10-12 members had almost two times the prevalence rate noted in smaller households of 1-3 members. This is consistent with the studies of Mendall et al,²⁶ Brown,⁵⁴ and Shi et al ¹²⁰ which showed that domestic overcrowding in childhood is a risk factor for H. pylori infection.

The association with the source of domestic water supply, type of convenience used in the home and the methods utilised for the disposal of domestic waste, were also associated with H. pylori seropositivity. The sources of drinking water used were mostly borehole and well water and this showed a higher risk of infection of H. pylori compared to pipe borne water. Also, use of pit latrine for faecal disposal was more risky than use of water closet toilet system. Open space disposal of household waste had a higher risk prediction than the public waste bin method. Several authors59,60,121-123 have associated high prevalence of H. pylori infection with poor sanitation, especially poor faecal disposal and contact with faecal matter. Considering drinking water, Bode et al ²² reported a significant seropositivity with drinking of non-municipal sources of water among pre-school aged children in Germany. It has also been reported that molecular methods have detected the presence of H. pylori DNA in river water, well water, as well as surface or shallow ground water.^125,126 An analysis of surface water and groundwater samples in Pennsylvania and Ohio, in which a microscopic technique that detected actively respiring microorganisms bound to monoclonal anti-H. pylori antibody was used, found sixty - one per cent of the samples to be contaminated with H.

pylori.¹²⁵ In contrast, a prospective cross-sectional epidemiologic survey of Czech children reported that domestic water sources and sanitary conditions had no association with the prevalence of H. pylori.¹²² This contrary observation could have been due to their sanitary conditions being uniformly of good standard as all the homes had hot water and inside toilet facilities.

The lack of association with the number siblings and number of persons sharing the same bed with the child, despite an association with increased total household population, was also observed by Shi et al ¹²¹ in a seroepidemiologic study in China.

This finding may suggest that other factors of close family contact, and not necessarily sleeping together, may be responsible for the association of high prevalence of this infection with increased household population. For instance, Brown⁵⁴ reported that the culture of sharing eating plates within the household and premastication of food for infant were found to be independent risk factors for this infection in children.

The data on household pets showed that keeping of pets is not very common among the subjects. Also, pet-keeping did not show any association with seropositivity to H.

pylori. This is consistent with the reports of other studies,^66,123 and supports the unlikely zoonotic transmission of H. pylori.

There was no association between H. pylori seropositivity and the type of school or the school environment of the subjects. This may suggest that school environment is not as important as the home environment in the epidemiology of H. pylori in the study locality.

Considering clinical symptoms in relation to H. pylori seropositivity, abdominal pains, and fetor oris showed a significant association and this association has been reported by other authors,^33,84,127 but only the duration of abdominal pains was significantly

associated with H. pylori seropositivity. The voluntary experimental ingestion of a pure culture of H. pylori by Marshall³³ is one of the studies which showed the association of H. pylori infection with fetor oris. In that study, the resolution of this symptom was also demonstrated with eradication therapy instituted against the organism. The association of H. pylori seropositivity with abdominal pains has also been seen in previous studies.^84,127 Malaty et al ⁸⁴ reported that the epidemiologic patterns of H.

pylori infections in symptomatic and asymptomatic children were significantly different, with younger children who had recurrent abdominal pains being more likely to be infected with H. pylori than older children with the same complaint. Again, Bittencourt et al,¹²⁸ stated that in the under seven years old school children, abdominal pain (albeit atypical and difficult to locate precisely),has been described in virtually all cases of H.

pylori infection.

Data from this study did not show any significant association between diarrhoea and H.

pylori seropositivity and Rothenbacher et al ⁷⁵ also reported a significantly less occurrence of acute diarrhoeal illness in H. pylori infected children and adults when compared to uninfected subjects. In contrast, some other studies had suggested that infection with H. pylori may predispose patients to other gastrointestinal infections.

Passaro et al ⁷⁴ reported that newly acquired H. pylori infection was followed with increased occurrence of diarrhoea. It is difficult to exclude other confounding factors which may explain the conflicting reports of these observational studies.

Considering the provisional diagnoses of the subjects, H. pylori seropositivity was significantly associated with gastroduodenal peptic ulcer disease which constituted 3.5% of the study population. Studies which have shown a marked reduction in the rate of ulcer relapse following eradication of H. pylori provide strong evidence implicating the organism in the pathogenesis of duodenal ulcer disease. Rauws and Tytgat¹²⁹ reported that in patients who are persistently H. pylori-positive, the rate of duodenal ulcer relapse may approach 80% at one year, while recurrence rate for patients rendered H. pylori -negative by antimicrobial therapy range between 0 – 12 %. Forbes et al ¹³⁰ documented an ulcer recurrent rate of only three per cent in duodenal ulcer patients followed up to seven years after H. pylori eradication. Also, duodenal ulcers have been reported to heal faster when treated with antibiotics and H2-blockers as compared to acid suppression alone.^130,13 These findings suggest a relationship between

H. pylori infection and gastroduodenal ulcer disease.