Send letter to journal:
Re: Association between SIDS and H pylori infection

This article in the November issue of the Archives of Diseases in Childhood on the association between sudden infant death syndrome (SIDS) and H pylori infection is confusing. I am very familiar with H pylori colonization in gastric biopsies in children and its association with gastritis, peptic ulcer,and gastric cancer. However, the implication that the organism can cause an unexpected infant death, eg, SIDS is shocking!

Unexplained infant deaths (SIDS) are "fertile" soil for speculators that apply new technology, eg, PCR (polymerase chain reaction) to uncover new associations. Unfortunately, these observations are not based on an infra-structure of knowledge of the causes of infant mortality. Caution needs to be exercised when applying PCR technology to postmortem tissue and "discovering" an answer. The possibilty of contamination is real, and in addition infants can die with something and not of it.

I would value a response from Drs Fleming, Blair, Bacon, and Berry who co-authored the CESDI study of SUDI.

Send letter to journal:
Re: SIDS and Helicobacter pylori

We were interested to read the article by Kerr et al[1] on the SIDS problem. With regard to the interesting results we would like to point out some related findings. As pointed out by Kerr et al, Helicobacter pylori (HP) is abundant in less advantageous parts of society where smoking is often frequent and sometimes where SIDS occur. The fact that smoking is often inversely related to HP’s ability to colonise and to be transmitted from mother to child[2] might indicate that it is sensitive to smoke itself or products thereof generated after smoke inhalation. It is interesting to note that endogenous products of smoke like nitrate and nitrite often inhibit bacterial growth.[3][4]

Furthermore, we have previously demonstrated that total breakdown of all ingested urea takes place in all normal infants without causing problems of ammonia intoxication.[4] This is in contrast to SIDS victims, most of whom have unmetabolised urea in their faeces.[5] Due to these related circumstances it may seem a little adventurous to suggest that ammonia produced by HP could cause death in SIDS.

Lars Wiklund
Gunnar Ronquist
Mary George

References

(1) Kerr JR, Al-Khattaf A, Barson AJ, Burnie JP. An association between sudden infant death syndrome (SIDS) and Helicobacter pylori infection. Arch Dis Child 2000;83:429-34.

(2) Brenner H, Bode G, Adler G, Rothenbacher D. Does maternal smoking hinder mother-child transmission of Helicobacter pylori infection? Epidemiology 2000;11:71-5.

(3) Goretski J, Zafiriou OC, Hollocher TC. Steady-state nitric oxide concentrations during denitrification. J Biol Chem 1990;265:11535-8.

(4) George M, Nord KE, Ronquist G, Hedenstierna G, Wiklund L. Faecal microflora and urease activity during the first six months of infancy. Ups J Med Sci 1996;101:233-50.

Send letter to journal:
Re: Control your controls and conclusions

Kerr and co-workers[1] in a retrospective study investigated formalin fixed, paraffin embedded tissues (stomach, trachea and lung) of 32 infants who had died of sudden infant death syndrome (SIDS), and eight control cases, with nested PCR and ELISA of the amplicons. A child was considered as infected with Helicobacter pylori if the optical density of the ELISA was above the mean value plus 2 SD obtained in the tissue of control infants.

The authors found that 28 of the 32 SIDS cases, but only one of the eight control cases fulfilled these criteria. They conclude from their results that H pylori infection may play a causative role in SIDS. We have serious doubts about their results and conclusions.

The control group was extremely small and we would expect most, if not all, of these eight infants to have received one or more antibiotics in high doses intravenously over several days prior to death, since the causes of death were bacterial meningitis, septicaemia, pneumonia, necrotising enterocolitis, ileal perforation, and prematurity. In contrast, few if any of the SIDS victims would have received intravenous antibiotics. Therefore, if control children had been colonised with H pylori, the bacteria may have been suppressed. These eight infants are certainly not appropriate controls for this kind of study.

Nested PCR is a very sensitive method with a high risk of false positive results caused by contamination. The applied ELISA is yet another amplifying method which also increases the risks of unspecific binding. Although the authors stated that they tried to minimise contamination, no precautions have been performed at the time of autopsy and preservation of the tissue due to the retrospective character of the study. Because of the low specificity of the methods used it is mandatory to prove the identity of the PCR amplicons as H pylori-specific by sequencing the products. Such confirmation is not reported in the paper. To demonstrate the specificity of their method, the authors also could have performed analyses on control tissues, eg, brain, unlikely to be H pylori infected even when other tissues were assessed as "positive".

The fact that H pylori was not demonstrated in the stomach, trachea, or lung by histology in any of the children must raise major concerns that the applied methods were not specific. Other methods for detection of H pylori infection like fluorescence in situ hybridisation (FISH) have not been applied.[2] The authors do not report whether any of the children had histological signs of acute or chronic gastritis, which is found even in young children with H pylori infection.[3] If the bacterial load was so small that neither the bacteria nor the associated inflammation could be detected by histology, it seems questionable that metabolic products produced by H pylori, eg, ammonia, may play a causative role in SIDS as suggested by the authors.

Finally, the authors mention that both H pylori infection and SIDS are more common in poor socio-economic populations but fail to provide any information on the ethnic and socio-economic background of their cases and control infants. From many epidemiologic studies and our own experience it seems extremely unlikely that 28 of 32 infants (87%) under 28 weeks of age are infected by H pylori in a country such as the UK unless these children are from immigrant groups. We are, for example, following a cohort of German children from birth with regular testing for H pylori infection by two non-invasive tests: the detection of H pylori antigen in stool (HpSA, Meridian Diagnostics, Cincinnati, USA) and the 13C-urea breath test corrected for estimated individual CO2 production rate.[4] Although a quarter of the children have at least one H pylori infected parent (positive serology and/or a positive 13C-urea breath test) only 1.5% of the children have positive tests during the first three years of age.

On publication, this paper was widely reported by the media. This is likely to result in considerable anxiety among young parents and pregnant women, feelings of guilt in parents of SIDS children and unjustified H pylori eradication therapy in asymptomatic children. Since neither the selection of the control group nor the methodology used is fully robust, this study does not permit valid conclusions on the association of H pylori infection with SIDS. We believe it is irresponsible to promote inconclusive results in the light of such inadequate data.

Sibylle Koletzko, MD
Nikolaos Konstantopoulos, MD
Dr v Haunersches Kinderspital
Ludwig-Maximillians-University
Munich, Germany

Norbert Lehn, MD
Institute of Medical Microbiology and Hygiene
University Regensburg
Regensburg, Germany

David Forman, PhD
Unit of Epidemiology and Health Service Research
The Medical School, University of Leeds,
Leeds, UK

References
(1) Kerr JG, Al-Khattaf A, Barson AJ, Burnie JP. An association between sudden infant death syndrome (SIDS) and Helicobacter pylori infection. Arch Dis Child 2000;83:429-34.

(2) Trebesius K, Panthel K, Strobel S, et al. Rapid and specific detection of Helicobacter pylori macrolide resistance in gastric tissue by fluorescent in situ hybridisation. Gut 2000;46:608-14.

(3) Radhakrishnan S, Al Nakib B, Kalaoui M, Patric J. Helicobacter pylori-associated gastritis in Kuwait: Endoscopy-based study in symptomatic and asymptomatic children. J Pediatr Gastroenterol Nutr 1993;16:126-9.

Send letter to journal:
Re: Association is not the same as causation

Send letter to journal:
Re: Death kisses for newborns?

Kerr et al (Arch Dis Child 2000;83:429-34) claim Helicobacter pylori (H pylori) as a potential etiologic factor in SIDS. Fatal systemic ammonia intoxication through hydrolysis of urea by H pylori-produced urease in the lungs and trachea following aspiration of gastric juice was proposed as a possible pathogenic pathway.

In general we cannot agree with this hypothesis. The molecular procedure (nested PCR and ELISA-based detection) used in this study could explain some inconsistent data (eg, H pylori DNA-detection in lungs or trachea but not in the stomach). Furthermore it is debatable whether haematoxylin & eosin (H&E) routine staining is an efficient method to visualize Helicobacter-like organisms. A Warthin-Starry-silver stain, modified Giemsa or immunochemistry would have been more advisable.

We also regret that no histopathological data were given which could have provided essential information about a possible infectious etiology. From our experience we observed that an acute H pylori infection always causes marked inflammatory changes of the gastric mucosa.

We also find that the negative control group was not a good reference, as this group did not comprise enough cases and was too heterogeneous (including 2 premature cases with apparently no normal environmental contact, one case with pneumonia (aspiration-pneumonia?)).

The discussion is totally speculative (eg, the role of interleukin-1 in H pylori infection: the main cytokines involved are (decreased production of) transfroming growth factor, (local production of) tumour necrosis factor (TNF), interleukin-2 and interleukin-8). From the data presented only the presence of H pylori-DNA in the respiratory system (some cases without infection of the gastric mucosa) can be claimed. All other conclusions are not substantiated and should be considered as speculative until further evidence is provided (eg, culturing of H pylori from tracheal or lung fluid).

Even if the presence of viable H pylori cells in the respiratory system can be established, some kind of experimental model should be used to establish H pylori as a causative agent in SIDS.

Recent findings established by the Children’s Hospital of Bamberg, Germany suggest a hypoplasia of the basilar artery as a more plausible explanation for SIDS. It has been shown that this anatomical defect can cause blockage of the cerebral blood circulation especially in the prone sleeping position when the head is turned aside. This hypoplasia can be detected by ultrasound. Data of this study[1] performed by the same hospital, seem to confirm this hypothesis. Among 3506 births over the last 2 years, 31 newborns (0.88%) could be identified with marked hypoplasia of the basilar artery, 6 of these newborns were considered as high risk cases (1.7‰). The babies were given a monitor and the parents were instructed in resuscitation. None of the children born and screened in the Children’s Hospital of Bamberg died from SIDS in the last years, whereas 2 babies not participating in the screening program died among 1130 house-born babies in the region of Bamberg (1.8‰). For statistical significance 5000 births are necessary; a number that will be reached end of the year 2001. Further reports are pending.

M Vieth*
D De Groote**
KH Deeg***
M Stolte*
G Seitz****

*Institute of Pathology, Klinikum Bayreuth, Germany
**Department of Pathology, Bacteriology and Avian Diseases, Ghent University, Merelbeke, Belgium
***Childrens clinic, Klinikum Bamberg, Germany
****Institute of Pathology, Klinikum Bamberg, Germany

Correspondence to:
Dr. med. M. Vieth
Institute of Pathology
Klinikum Bayreuth
Preuschwitzerstr. 101
95445 Bayreuth
Tel: ++ 49 921 400 5620
Fax: ++49 921 400 5609
e-mail: vieth.lkpathol@uni-bayreuth.de

Send letter to journal:
Re: A link between dwelling crowding, Helicobacter pylori infection and sudden infant death syndrome

Kerr et al[1] report a highly significant association between Helicobacter pylori infection and sudden infant death syndrome (SIDS). This finding raises the possibility of (and a plausible mechanism for) a link between dwelling crowding and SIDS, as there are a number of studies that have documented a strong relation between dwelling crowding and H pylori infection (eg, Fall et al[2]; Webb et al[3]). Close person to person contact and increased exposure to the infective agent is a likely cause of this relationship. Dwelling crowding has also been associated with increased passive exposure to tobacco smoke, and this, coupled with parental smoking being strongly associated with SIDS,[4] provides yet another clear link between dwelling crowding and SIDS.

There are likely to be many causes of dwelling crowding. It has often been associated with low socioeconomic status, but the study by Elitsur et al[5] suggests that there may be a direct link between crowding and H pylori infection, which is independent of socioeconomic status. SIDS has also been associated with lower environmental temperature (ie, winter), and it is possible that the increase in SIDS rate during winter is in part relate to the increased dwelling crowding during such times.

Very few studies have examined the links between dwelling crowding and SIDS. One recently published study found only a non-significant increase in relative risk for SIDS associated with dwelling crowding.[6] Given the importance of SIDS and the growing body of evidence suggesting H pylori as a cause of SIDS, it would be pertinent for future studies to consider dwelling crowding in more detail.

References
(1) Kerr JR, Al-Khattaf A, Barson AJ, Burnie JP. An association between sudden infant death syndrome (SIDS) and Helicobacter pylori infection. Arch Dis Child 2000;83:429-34.

(2) Fall CHD, Goggin PM, Hawtin P, Fine D, Duggleby S. Growth in infancy, infant feeding, childhood living conditions, and Helicobacter pylori infection at age 70. Arch Dis Child 1997;77:310-14.

(3) Webb PM, Knight T, Greaves S, et al. Relation between infection with Helicobacter pylori and living conditions in childhood: evidence for person to person transmission in early life. BMJ 1994;308:750-4.

(4) Henderson-Smart DJ, Ponsonby AL, Murphy E. Reducing the risk of sudden infant death syndrome: a review of the scientific literature. J Paediatr Child Health 1998;34:213-19.

(5) Elitsur Y, Short JP, Neace C. Prevalence of Helicobacter pylori infection in children from urban and rural West Virginia. Dig Dis Sci 1998;43:773-8.

Send letter to journal:
Re: Does Helicobacter pylori have a role in sudden infant death syndrome?

Kerr et al[1] report an association between sudden infant death syndrome (SIDS) and Helicobacter pylori infection. In 32 SIDS cases aged up to 28 weeks old the H pylori ureC gene was amplified from the stomachs of 15, from the trachea of 19, and from the lungs of 16. The H pylori cagA gene was amplified from the stomachs of 13 (of which seven were positive for the ureC gene), from the trachea of 20, and the lungs of 20 (of which 14 were positive for the ureC gene). Amplified DNA was detected semiquantatively using an ELISA, with a cutoff value calculated from the mean of eight controls. The authors offered little explanation for the discordant detection of H pylori DNA between the two PCR assays used. It may be appropriate to compare the prevalence of H pylori in SIDS and controls, but inappropriate to make these two groups the basis for defining cutoffs for an H pylori assay.

The presence of H pylori DNA does not itself imply infection and no visible bacteria were observed in any tissue sections. H pylori can be acquired early in life[2] probably from other members of the family. Infection has only previously been detected in the microenvironment of the gastric mucosa and its presence is closely related to socioeconomic status,[3] as is SIDS. No details of the socioeconomic status of the infants from whom tissues were obtained, nor details of familial contact were given. Four of the controls died under eight weeks of age from what could possibly be neonatal complications and no details of whether they had been discharged home were provided.

The authors propose that primary gastric infection and subsequent aspiration into the lungs led to lethal production of ammonia in infants as young as two weeks of age. It is difficult to imagine that an organism specifically adapted to the microaerophilic and acidic conditions of the gastric mucosa thriving well enough in the lungs to produce toxic amounts of ammonia in infants that presumably had normal livers, particularly when no organisms were visible on histology.

This interesting report could well describe a proxy for the already widely known association between H pylori and poor socio-economic status. Arguing that the discordant presence of H pylori DNA in various organs of SIDS cases represents causation is premature, but warrants further investigation.

CP Doherty*
WG MacKay*
AJ Shepherd†
CL Williams**
LT Weaver*

*Department of Child Health, University of Glasgow
†Department of Nursing and Midwifery, Stirling University
**Department of Microbiology, Royal Alexandra Hospital, Paisley
Scotland

References
(1) Kerr JR, Al-Khattaf A, Barson J, Burnie JP. An association between sudden infant death syndrome (SIDS) and Helicobacter pylori infection. Arch Dis Child 2000;83:429-34.

(2) Thomas JE, Dale A, Harding M, Coward WA, Cole TJ, Weaver LT. Helicobacter pylori colonisation in early life. Pediatr Res 1999;45:218-23.

Send letter to journal:
Re: The link between H pylori and SIDS - is it proven to be causal?

I am writing to make my observations on the article by Kerr et al[1] suggesting a direct causal link between H pylori and SIDS. I query that the link is indeed proven on 2 counts:

(1) When infants with SIDS are discovered in extremis by their parents/carers they invariably receive mouth to mouth resuscitation till emergency services are on the scene and take over. Saliva is a recognised mode of transmission of H pylori.[2][3] It has been shown that H pylori can be transmitted through mouth to mouth resuscitaion.[4] Given the sensitivity of the PCR technique,is it then possible that the positive results are actually from the oral/pharyngeal secretions of carers transmitting the organism during the resuscitative attempts? This is all the more so considering the common epidemiological features of SIDS and H pylori.

(2) The control infants had died of various causes - but mainly infection related. It is very likely that they were treated with antibiotics. Could this have reduced the possibility of getting a positive result for H pylori? Also these children are unlikely to have received mouth to mouth resuscitation.

Dr V S Sankar
Consultant Paediatrician
Fairfield General Hospital
Bury, Greater Manchester, UK

References
(1) Kerr JR, et al. An association between SIDS and H pylori infection. Arch Dis Child 2000;83:429-34.

(2) Parsonnet J, et al. Fecal and oral shedding of H pylori from healthy infected adults JAMA 1999;282:2240-5.

(3) Madinier IM, et al. Oral carriage of H pylori: A review J Periodontol 1997;68:2-6.

Send letter to journal:
Re: Further information on SIDS and Helicobacter pylori infection - Authors' response

Dear Editor,

Following the publication of our paper,[1] we would like to thank the authors for their comments and respond to them to clarify our study methodology, interpretation of the data, the impact of the media and on the directions of future work in this area.

The possibility of PCR contamination has been suggested by Franciosi and Koletzko and we agree that this is a potential problem in studies of this type. We guarded against this by utilisation of separate laboratory areas and pipettes for pre-PCR, PCR and post-PCR stages of the procedure, use of sterile bunged pipette tips and inoculation of the positive control as a last step in the pre-PCR preparation. In each run, we used sterile distilled water and DNA extract from human ureter as negative controls, and we examined samples in duplicate. Throughout our study, duplicated samples consistently gave concordant results, and negative controls were consistently negative.[1]

Dr Koletzko suggests that the two separate nested PCR-ELISAs utilised in our study may have doubtful specificity as we did not sequence the products. We agree that amplicon sequencing is desirable not only to ensure specificity but in the present context would also provide additional data on the molecular epidemiology of the cagA gene which was detected in these cases. We believe our assays to be specific. For example, the binding of oligonucleotides of 20 or more bases to template DNA at 55ºC has been shown to be 100% specific. And in one of our PCR- ELISAs, there were five such interactions.[1] We agree with Dr Koletzko and other workers that it would be valuable to test other tissues from the same patients by the same method.

Regarding our controls, use of these cases is quite appropriate and to illustrate this we include additional relevant data in the Table. Dr Vieth says our controls have had no normal environmental contact, however, five of these eight had spent time in the home environment since birth (Table). Regarding antibiotic treatment, only one control had received antibiotics for more than one day prior to death, and this is the case in which H pylori was detected (Table).

Information on antibiotic exposure, environmental exposure and PCR-ELISA testing for H pylori ureC and cagA genes in the stomach, trachea and lung of control cases used in the study: An association between sudden infant death syndrome (SIDS) and Helicobacter pylori infection.[1] Results of PCR-ELISA testing is expressed as optical density. Those specimens with a cut-off value greater than or equal to the mean plus two times the standard deviation of these controls (designated negative) are marked with an asterisk.
 

C1, control case number 1; AM, ante-mortem; PM, post-mortem; NT, not tested

Dr Koletzko states that "the fact that H pylori was not demonstrated in the stomach, trachea or lung by histology in any of the children must raise major concerns that the applied methods were not specific." However, as pointed out by Dr Vieth, haematoxylin and eosin staining, although a routinely used stain in histopathology practice, may not be optimal for microscopic visualisation of H pylori.

In response to Dr Vieth's claim that our suggested role for interleukin-1beta (IL-1beta) in H pylori infection is "totally speculative", we would like to point out that these mechanisms have been demonstrated in an animal model.[2][3] Also, proteins of H pylori are known to activate macrophages leading to production of IL-1beta[4-6] which is known to inhibit acid secretion by parietal cells and may actually be the most potent inhibitor of acid secretion discovered to date.[7] IL-1beta gene polymorphisms associated with increased IL-1beta production have recently been associated with an increased risk of gastric cancer.[8] In addition, systemic and mucosal humoral recognition of the cagA protein has been linked with peptic ulceration,[9][10] duodenal ulcer patients may more frequently harbour cagA+ H pylori strains,[6][10] and it has been demonstrated that infection with cagA+ as compared with cagA- strains is associated with increased transcription of IL-1beta.[6] It is therefore interesting that 25 of 28 cases of H pylori-associated SIDS in our study had a detectable cagA gene in their tissues,[1] which may provide further support for the proposed pathogenesis of H pylori in SIDS and a contributory role for IL-1beta.[11]

Dr Paul Beggs from Macquarie University in Australia points out the link between dwelling crowding and H pylori infection,[12][13] which has been shown to be independent of socioeconomic status,[14] and the need for research on the possible link between dwelling crowding and SIDS. We agree that "given the importance of SIDS and the growing body of evidence suggesting H pylori as a cause of SIDS, it would be pertinent for future studies to consider dwelling crowding in more detail."

We feel that Wiklund and colleagues and MacKay and colleagues have misunderstood the proposed hypothesis. Wiklund states that total breakdown of ingested urea occurs in all normal infants without ammonia intoxication and that SIDS victims have undigested urea in their faeces. MacKay states that "it is difficult to imagine that an organism specifically adapted to the microaerophilic and acidic conditions of the gastric mucosa thriving well enough in the lung to produce toxic amounts of ammonia in infants that presumably had normal livers." To reiterate, there are two parts to the hypothesis. First, interleukin-1b production in the H pylori-infected stomach, and second, supply of ammonia to the systemic circulation11 (and not the hepatic circulation as MacKay implies). Therefore, faecal urea content is irrelevant and so is ammonia produced in the stomach as this will be detoxified by the liver.

Regarding comments in the media, these are clearly not under our control and we have always stated that our findings are preliminary and require confirmation.

In conclusion, we would encourage researchers to repeat our studies and those of Pattison and colleagues[15-17] in order to clarify the proposed role of H pylori in SIDS. In the meantime, we re-emphasise accepted measures to reduce mortality from SIDS and suggest the following additional precautions, all of which constitute good personal hygiene and are therefore advisable even in the absence of such a link. First, to prevent the transfer of saliva from the mouths of carers to babies. Second, prompt disposal of vomitus, decontamination of soiled surfaces, and washing of soiled clothes/bedclothes, followed by hand washing, in order to minimise transmission to the baby via the gastro-oral route. Third, good general hand and personal hygiene. In addition, parents should be reassured that they do not need to do anything more than the above at present.

JR Kerr
JP Burnie
Infectious Diseases Research Group
The University of Manchester

AJ Barson
Department of Paediatric Pathology
The University of Manchester
Clinical Sciences Building
Manchester Royal Infirmary
Oxford Road, Manchester M13 9WL, UK

References
(1) Kerr JR, Al-Khattaf A, Barson AJ, Burnie JP. An association between sudden infant death syndrome (SIDS) and Helicobacter pylori infection. Arch Dis Child 2000;83:429-34.

(2) Pattison CP, Marshall BJ, Scott LW, Herndon B, Willsie SK. Proposed link between Helicobacter pylori and sudden infant death syndrome (SIDS): possible pathogenic mechanisms in an animal model. I. Effects of intratracheal urease. Gastroenterology 1998;114:G3689.

(3) Pattison CP, Scott LW, Herndon B, Willsie SK. Proposed link between Helicobacter pylori and SIDS: possible pathogenic mechanisms in an animal model. II. Effects of intratracheal urease after pretreatment with intravenous IL-1 beta. Gastroenterology 1998;114:G3690.

(4) Noach LA, Bosma B, Jansen J, Hoek FJ, vanDeventer SJ, Tytgat GN. Mucosal tumour necrosis factor-a, interleukin-1b, and interleukin-8 production in patients with Helicobacter pylori infection. Scand J Gastroenterol 1994;29:425-9.

(5) Yamaoka Y, Kita M, Kodama T, Sawai N, Kashima K, Imanishi J. Expression of cytokine mRNA in gastric mucosa with Helicobacter pylori infection. Scand J Gastroenterology 1995;30:1153-9.

(6) Peek RM, Miller GG, Tham KT, et al. Heightened inflammatory response and cytokine expression in vivo to cagA+ Helicobacter pylori strains. Lab Invest 1995;73:742-5.

(7) Wallace JL, Cucala M, Mugridge, Parente L, Secretagogue-specific effects of interleukin-1 on gastric acid secretion. Am J Physiol 1991;261:G559-64.

(8) El-Omar EM, Carrington M, Chow WH, et al. Interleukin-1 polymorphisms associated with increased risk of gastric cancer. Nature 2000;404:398-402.

(9) Crabtree JE, Taylor JD, Wyatt JI, et al. Mucosal IgA recognition of Helicobacter pylori 120 kDa protein, peptic ulceration and gastric pathology. Lancet 1991;338:332-5.

(10) Cover TL, Glupczynski Y, Lage AP, et al. Serologic detection of infection with cagA+ Helicobacter pylori infections. J Clin Microbiol 1995;33:1496-500.

(11) Pattison CP, Marshall BJ. Proposed link between Helicobacter pylori and sudden infant death syndrome. Med Hypotheses 1997:49:365-9.

(12) Fall CHD, Goggin PM, Hawtin P, Fine D, Duggleby S. Growth in infancy, infant feeding, childhood living conditions, and Helicobacter pylori infection at age 70. Arch Dis Child 1997;77:310-14.

(13) Webb PM, Knoght T, Greaves S, et al. Relation between infection with Helicobacter pylori and living conditions in childhood: evidence for person to person transmission in early life. BMJ 1994;308:750-4.

(14) Elitsur Y, Short JP, Neace C. Prevalence of Helicobacter pylori infection in children from urban and rural West Virginia. Dig Dis Sci 1998;43:773-8.

(15) Pattison CP, Marshall BJ, Young TW, Vergara GG. Is Helicobacter pylori the missing link for sudden infant death syndrome? Gastroenterology 1997;112(4SS):A254.

(16) Pattison CP, Vergara GG, Young TW, Smith GP. Prevalence of Helicobacter pylori in sudden infant death syndrome. Gastroenterology 1998;114:G3688.

Send letter to journal:
Re: The need for further evidence for the proposed role of Helicobacter pylori in SIDS

We read with interest the article by Kerr et al "An association between sudden infant death syndrome (SIDS) and Helicobacter pylori infection". While the proportion of samples positive for H pylori DNA were significantly higher in the SIDS group compared with the control group, the findings need to be interpreted with caution.

PCR is a useful tool for detection of DNA. It is, however, evidence that the DNA of the organism is present, not evidence that the organisms were alive or caused disease. Culture, microscopy, serological evidence or histological evidence of inflammatory or immune responses are needed to support the hypothesis that the bacteria were involved with pathological processes, not just transient contamination of the infant with DNA from non-viable bacteria.

There are several points that detract from the paper:

In relation to the findings reported:
1. Only the PCR assays provided positive evidence. In contrast to other studies reported as abstracts, microscopic examination of the stained sections did not find any evidence of H pylori. This discrepancy needs to be explained. There were no serological data to support the PCR findings and no data from histological examinations to provide evidence that the bacteria were causing infection or that inflammatory responses had been elicited.
2. The proportion of PCR positive samples among SIDS infants (88%) was significantly larger than that among controls (12.5%). The possibility of contamination was not addressed for SIDS or the positive control case. There was no demonstration by molecular methods that the DNA detected was from different strains. H pylori strains show great genetic variability and previous studies demonstrated that most individuals carry unique strains. Isolates from different individuals that appear to be genetically identical are those obtained only from close contacts, usually within a family.

The interpretation of the epidemiological data for H pylori and socioeconomic factors was not assessed in relation to incidence of SIDS among different ethnic groups. In Britain, white families in lower socioeconomic groups have more evidence of H pylori infections and more SIDS. If the data for incidence of infections with H pylori is assessed for ethnic groups and SIDS, this parallel breaks down. The incidence of seropositivity for H pylori among Bangladeshi women in the UK ranges from 66% among women born abroad to 81% among women born in the UK [Banatvala et al, 1995]; however, the incidence of SIDS in Bangladeshi families was the lowest in Britain (0.3%) [Balarajan et al, 1989]. A similar trend was observed in the United States; seropositivity for H pylori is 61% among Hispanics and 26.2% among non-Hispanic whites [Everhart et al, 2000]. In the paper quoted in the manuscript, the incidence of seropositivity was similar for Hispanic and black groups and both were significantly higher than that of non-Hispanic whites [Malaty et al, 1992]. The SIDS rates per 100,000 for US populations were 5.1 for blacks, 1.3 for Hispanics and 1.2 for non-Hispanic whites [Nelson, 1996]. This evidence questions the assumptions made by the authors.

While there is increasing evidence for other hypotheses that SIDS might be triggered by inflammatory responses to infection [Vege and Rognum, 1999; Raza and Blackwell, 1999; Forsyth, 1999], there is no physiological or histological evidence to support the hypothesis that urease in the lung of the infants is causing increased levels of ammonia in the blood (see detailed assessment of pathology of SIDS in relation to this hypothesis below). Animal models [Pattison et al, 1998a,b] do not reflect the combination of genetic, environmental and developmental factors associated with SIDS, and results from animal studies must be interpreted with extreme caution when extrapolated to the human infant.

H pylori infection does not fit the common bacterial hypothesis, a mathematical model which accurately predicted the age range for SIDS [Morris, 1999]. According to the model, 50% of infants should acquire the bacteria during the first 50 days of life. While 19% of Gambian children were positive for the C13 urea breath test by 3 months of age [Thomas et al, 1999], in industrialised countries the evidence is that H pylori infection in infants under 1 year of age is much lower. Among 67 Belgian children born to seropositive mothers, only 1 (1.5%) had a positive breath test by the age of 12-15 months. [Blecker et al, 1994]. Among Finnish children 10.6% had IgG to H pylori at birth, but the antibodies disappeared in all but one child before the age of 7 months and there were no seroconversions in these children. The Finnish study concluded that maternal seropositivity is not a straight forward risk factor for acquiring H pylori infection [Ashorn et al, 1996].

The oral/oral route of transmission is suggested to be the route by which infants acquire H. pylori, mainly by vomit [Leung et al, 1999]. H pylori has been cultured from one of four vomit samples from children and detected by PCR in two of four culture negative samples. There is much stronger direct (culture) evidence for transmission from mother to child of other bacterial species implicated in SIDS [Blackwell et al, 1999a].

The pathogenic mechanism proposed for the role of ammonia cannot be substantiated by the available evidence:
a. There are no acute changes in the upper respiratory tree or lungs consistent with inflammatory responses to H pylori.
b. The presence of ammonia in the lower respiratory tree would initiate a bronchospasm which should produce clinical features such as wheezing which has not been reported by parents of SIDS infants. This type of reaction should be demonstrable histologically by muscular, glandular and secretory changes identified by microscopy.
c. If ammonia is present in excess in the blood as a proximate cause of death, this should be demonstrable in blood samples and vitreous fluid, and there is no evidence for this.
d. The liver in SIDS cases shows no abnormality and had it been acutely affected by an influx of ammonia, there should be changes.
e. Ammonia in excess leads to cerebral changes of an acute type and none have been demonstrated.
f. If the ammonia is postulated as a cause of petechiae in the lungs due to local damage, this does not account for the presence of petechiae in the thymus and pericardium.

There is evidence to explain how risk factors could contribute to susceptibility of infants to infectious agents to triggering the series of events leading to SIDS [Blackwell et al, 1999b]; however, that presented for H pylori needs to be substantiated by more than one method and testable hypotheses proposed to explain how these bacteria might contribute to the series of events that lead to SIDS.

CC Blackwell, PhD, FRCPath DSc
DM Weir, MD FRCPE
A Busuttil, MD, DMJ FRCPath FRCP

References
Ashorn M, Miettinen A, Ruuska T, et al. Seroepidemiological study of Helicobacter pylori infection in infancy. Arch Dis Child Fetal Neonatal Ed 1996; F141-2.

Balarajan R, Raleigh VS, Botting B. Sudden infant death syndrome and post neonatal mortality in immigrants in England and Wales. BMJ 1989;298: 716-20.

Banatvala N, Clements L, Abdi Y, et al. Migration and Helicobacter pylori seroprevalence: Bangladeshi migrants in the UK. J Infect 1995;31:133-5.

Blackwell CC, Mackenzie DAC, James VS, et al. Toxigenic bacteria and SIDS: nasopharyngeal flora in the first year of life. FEMS Immunol Med Microbiol 1999;25:51-8.

Blackwell CC, Weir DM, Busuttil A. Infection, inflammation and sleep: more pieces to the puzzle of sudden infant death syndrome (SIDS). APMIS 1999;107:455-73.

Blecker U, Lanciers S, Keppens E, Vandenplas Y. Evolution of Helicobacter pylori positivity in infants born from positive mothers. J Pediatr Gastorenterol Nutr 1994;19:87-90.

Everhart JE, Kruszon-Moran D, Perez-Perez GI, et al. Seroprevalence and ethnic differences in Helicobacter pylori infection among adults in the United States. J Infect Dis 2000;181:1359-63.

Forsyth KD. Immune and inflammatory responses in sudden infant death syndrome. FEMS Immunol Med Microbiol 1999;25:79-84.

Leung WK, Siu KL, Kwok CK, et al. Isolation of Helicobacter pylori from vomitus in children and its implication in gastro-oral transmission. Am J Gastroenterol 1999;94:2881-4.

Malaty HM, Evans DG, Evans DJ, Graham, DY. Helicobacter pylori in Hispanics: comparison with blacks and whites of similar age and socioeconomic class. Gastroenterology 1992;103:813-16.

Morris JA. The common bacterial toxins hypothesis of sudden infant death syndrome. FEMS Immunol Med Microbiol 1999;25:11-18.

Nelson T. Sudden infant death syndrome and child care practices. Lammar Offset Pringing, Ltd. Hong Kong, 1996.

Pattison CP, Marshall BJ, Scott LW, et al. Proposwed link between Helicobacter pylori and sudden infant death syndrome (SIDS): possible pathogenic mechanisms in an animal model. I. Effects of intratracheal urease. Gastroenterology 1998;114:G3689.

Pattison CP, Scott LW, Herndon B, Willsie SK. Proposed link between Helicobacer pylori and SIDS: possible pathogenic mechanisms in an animal model. II. Effects of intratracheal urease after pretreatment with intravenous IL-1b. Ibid.

Raza MW, Blackwell CC. Sudden infant death syndrome: virus infections and cytokines FEMS Immunol Med Microbiol 1999;25: 85-96. Gastroenterology 1998;114:G3690.

Thomas JF, Dale A, Harding M, et al. Helicobacter pylori colonization in early life. Pediatric Res 1999;45:218-23.

Send letter to journal:
Re: H pylori and SIDS

Dear Editor,

A recent paper by Kerr et al reported an association between H pylori and sudden infant death syndrome (SIDS). We have reviewed their data and believe that the methods used may have led to incorrect conclusions.

Kerr et al examined retrospective material from 32 cases of SIDS infants and 8 non-SIDS controls. They used nested PCR followed by an ELISA detection step which would have made their method exquisitely sensitive. Consistent with this, no other method was able to confirm that H pylori was actually present. Instead, Kerr et al used a relative increase of “H pylori signal” above that of the mean +2SD for a control group, as an indicator of H pylori presence. This prompted us to more carefully consider the appropriateness of their control and patient groups.

Since ethnicity and socioeconomic details of the SIDS infants were not given, we could not confirm that these matched the control infants. We also noted important clinical details of the controls which could make them inappropriate. It appears that most of the controls would have had very little bacterial contamination of the PCR specimens because they died in hospital while on antibiotic therapy for sepsis, or were deceased very soon after premature birth. In addition, they might have been transferred to refrigeration very soon after death. SIDS infants however, probably died at home, many hours before being refrigerated.

Finally, since H pylori is a gastric organism, it was surprising to find the bacterium in lung or trachea of eight patients (ureC gene) or six patients (cagA gene) in whom gastric specimens were negative.

Since Kerr's paper was widely reported in the media, we believe that it needs to be stated that the case for H pylori as a cause of SIDS is certainly unproven and is in quite considerable doubt.

Send letter to journal:
Re: H pylori & SIDS

We read with great interest the paper by Kerr et al on the association between Helicobacter pylori infection and sudden infant death syndrome (SIDS).[1] However, we cannot agree with the speculation the authors made.

Recently, we have performed a similar retrospective analysis of 9 cases of SIDS and 8 controls collected in our hospital over the past two years. Controls were selected from infants with known cause of deaths including congenital malformation, infection, metabolic disease and drug intoxication (see Table).

The formalin-fixed and paraffin-embedded stomach, trachea and lung specimens obtained during post-mortem examination were retrieved. Initial histological examination was performed by an experienced pathologist to look for any evidence of H pylori colonization in these specimens. In addition, we used three different PCR assays that amplify two regions of the ureB gene[2] [3] and the cagA gene[4] to detect the presence of H pylori DNA in these samples.

Histological examination failed to show any Helicobacter like organism in these samples. Moreover, despite using three different sensitive PCR assay, we failed to demonstrate the presence of H pylori DNA in the stomach, lung or trachea of the SIDS and control patients.

Viable H pylori has recently been recovered from the vomitus of infected children and adult.[3] Conceivably, it could lead to silent aspiration of gastric contents into the lung and result in bronchopneumonia. However, the failure to detect the organism in the stomach, trachea and lung specimens, together with the absence of features to suggest aspiration pneumonia as the cause of death in these infants, argue against the validity of this speculation. With the high prevalence of H pylori infection in Chinese, one would expect a parallel high incidence of SIDS in our ethnic group, which does not fit into any epidemiological observations. Taken together, the significance of H pylori as a cause of SIDS is highly questionable.

Wai K Leung
Jun Yu
K F To
Department of Medicine & Therapeutics and Department of Anatomical & Cellular Pathology
Chinese University of Hong Kong
Shatin, Hong Kong

References
(1) Kerr JR, Al-Khattaf A, Barson AJ, Burnie JP. An association between sudden infant death syndrome (SIDS) and Helicobacter pylori infection. Arch Dis Child 2000;83:429-34.

(2) Leung WK, Sung JJY, Ling TKW, Siu KLK, Cheng AFB. Does the use of chopsticks for eating transmit Helicobacter pylori? Lancet 1997;350:31.

(3) Leung WK, Siu KLK, Kwok CKL, Chan SY, Sung R, Sung JJY. Isolation of Helicobacter pylori from vomitus in children and its implication in gastro-oral transmission. Am J Gastroenterol 1999;94:2881-4.

Send letter to journal:
Re: On sudden infant death syndrome (SIDS) and Helicobacter pylori infection - Authors' response

Dear Editor,

At present, we do not understand the pathogenesis of sudden infant death syndrome (SIDS), however, it is accepted to be a multifactorial disease for which certain risk factors have been identified. Various theories have been developed to explain the existence of these risk factors.

Blackwell reminds us of the accepted fact that PCR detects DNA from both live and dead organisms, but her phrase "transient contamination of the infant with DNA from non-viable bacteria" seems inappropriate. The detection of Helicobacter pylori DNA in the trachea and lung of such babies is a finding of particular importance both for our understanding of the pathogenesis of SIDS and for our understanding of the pathogenesis and epidemiology of H pylori infection in infants.

The study by Kerr et al (2000) demonstrated H pylori DNA in the stomach, trachea and lung tissues of SIDS cases, but did not visualise bacteria at these sites. As stated in the paper and by several other authors, the study utilised haematoxylin & eosin staining, a sub-optimal methodology for visualisation of gastric bacteria. Other studies have demonstrated inflammatory changes in both antrum and trachea of H pylori-PCR positive SIDS cases (Pattison et al, 1998)

Genetic subtyping would be valuable as suggested, but not essential, as the PCR-ELISA utilised was specific, tests were performed in duplicate and positive and negative controls consistently gave expected results (Kerr et al, 2000).

Our hypothesis is that H pylori infection accounts for a proportion of cases of SIDS. Blackwell cites several epidemiological papers, stating that they argue against this hypothesis, but she does not state how exactly she considers that they do argue against it. Epidemiological data for H pylori and socioeconomic factors in various ethnic groups are not clear cut and are incomplete. Such factors as prevalence of bottle feeding, parental smoking, family size, adherence to supine sleep position, etc, may explain differences of SIDS incidence in various ethnic groups. Blackwell's use of data regarding breath testing in children aged 12-15 months is in contrast to the finding of 44% H pylori- positivity by 13C-urea breath testing of 2 year olds in childcare centres serving low socioeconomic groups in Houston, Texas (Malaty et al, 1996).

Blackwell reminds us of the accepted fact that animal work is not directly applicable to events in the human infant. But, it is relevant. The proposed hypothesis cannot be verified in the human infant, but this should not be taken as evidence that it does not account for infant mortality.

The common bacterial hypothesis (Morris, 1999) has been useful for studies of other bacteria (Blackwell et al, 1999), but is not a basis for rejection of conflicting data. While other bacteria may be readily transmitted from mother to infant (Blackwell et al, 1999), none has been consistently linked with SIDS, and transmission efficiency does not equate with pathogenicity in a particular setting.

There are three proposed routes of transmission of H pylori; oral- oral, gastro-oral and faecal-oral (Feldman et al, 1998). Blackwell has misunderstood these, as she refers to transmission by vomit as "oral-oral", when this is actually gastro-oral (Feldman et al, 1998). The transmission of H pylori is more complex than that of other oral bacteria.

The proposed pathogenesis of the involvement of H pylori in SIDS is that death may occur as a result of one or both of two events (Pattison et al, 1997), both of which have been demonstrated in a rat model (Pattison et al, 1998b; 1998c). First, H pylori produces large amounts of urease, which will be fully active in the neutral pH of the H pylori-infected stomach (Mobley et al, 1988). Therefore, aspiration of this gastric juice may lead to large amounts of urease in the alveolae in close proximity to plasma urea. In this setting, urea hydrolysis may lead to ammonia production and supply directly to the systemic circulation where it cannot be detoxified by the liver (Pattison et al, 1998b); unlike the case of ammonia production within the gastric mucosa. Intravenous administration of ammonia is known to be fatal (Fitzgerald et al, 1950). Second, IL-1b produced in the H pylori-infected gastric mucosa and may lead to fever, immune activation and increased deep sleep, which in combination with supply of ammonia to the systemic circulation may be lethal (Pattison et al, 1998c). Increased production of IL-1b alone as a result of gastric H pylori infection may predispose to the development of SIDS due to other factors (Kerr, 1998).

Blackwell states that the proposed pathogenesis cannot be substantiated due to the following:
(a) "There is no inflammation in the lungs of SIDS cases". This is not true; mild inflammation of the upper respiratory tract is a recurrent, although not invariable, finding in SIDS (Guntheroth, 1995).
(b) "Ammonia in the lower respiratory tract would cause bronchospasm and wheezing which has not been reported by SIDS parents". In animal studies (not yet published as a full paper), bronchospasm was suggested by progressively less bronchoalveolar lavage (BAL) fluid return after sequential doses of intratracheal urease (Pattison et al, 1998b). Since parents are invariably absent at the time of death, it would be unlikely that wheezing would be detected. "If bronchospasm occurs, this should be demonstrable histologically". Findings of relevance in SIDS include intrathoracic petechiae, patchy pulmonary oedema, emphysema, and increased muscle mass in pulmonary arteries (Guntheroth, 1995), although these are not invariable findings.
(c) "If ammonia accounts for death, this should be demonstrable in blood and vitreous". Our hypothesis is supported by intratracheal urease adminisatration to rats which caused increased ammonia in BAL fluid although this was not accompanied by significantly increased serum ammonia (Pattison et al, 1998b). The physiological effects of pre-treatment with IL-1b could not be clearly defined (Pattison et al, 1998c).
(d) "The liver should be affected by hyperammonaemia and it is not in SIDS". Blackwell has misunderstood our hypothesis. First, interleukin-1b production in the H pylori-infected stomach, and second, aspiration of urease into the lung and supply of ammonia to the systemic circulation (and not the hepatic circulation as Blackwell implies) (Pattison et al, 1997).
(e) "The brain should be affected by hyperammonaemia and it is not in SIDS". If our hypothesis is correct, then the terminal event, involving hyperammonaemia in the systemic circulation is an acute and rapidly fatal occurrence, which may not result in brain pathology.
(f) We do not understand this point.

Marshall's views on controls used in the original paper (Kerr et al, 2000) do not take account of further information provided at the request of other authors (Kerr et al, 2001) which show that of eight controls used, five had an exposure to the home environment of more than one month.

Marshall states that H pylori is a gastric organism and that it is surprising to find evidence of infection in lung and trachea. However, H pylori has been detected at other sites, for example, the respiratory tract of both SIDS victims (Pattison et al, 1998a) and intubated adults (Mitz et al, 1993), and in the liver of patients with primary sclerosing cholangitis and primary biliary cirrhosis (Nilsson et al, 2000).

The pathogenesis of SIDS is accepted to be multifactorial, and therefore, small studies with a negative association between H pylori and SIDS, such as that of Leung and colleagues, are to be expected.

Jonathan R Kerr
Infectious Diseases Research Group
University of Manchester, Manchester, UK

C Phillip Pattison
Department of Medicine
University of Missouri at Kansas City School of Medicine
Kansas City, Missouri, USA

References
Blackwell CC, Mackenzie DAC, James VS, et al. Toxigenic bacteria and SIDS: nasopharyngeal flora in the first year of life. FEMS Immunol Med Microbiol 1999;25:51-8.

Feldman RA, Eccersley AJP, Hardie JM. Epidemiology of Helicobacter pylori: acquisition, transmission, population prevalence and disease-to- infection ratio. Br Med Bull 1998;54:39-53.

Fitzgerald O, Murphy P. Studies on the physiological chamistry and clinical significance of urease and urea with special reference to the stomach. Ir J Med Sci 1950;292:96-159.

Guntheroth WG. Cribdeath - the sudden infant death syndrome. Armonk, NY: Futura, 1995:91.

Kerr JR. Sudden infant death syndrome, long QT interval and Helicobacter pylori. J Clin Pathol 1998;332:943-4.

Kerr JR, Al-Khattaf A, Barson AJ, Burnie JP. An association between sudden infant death syndrome (SIDS) and Helicobacter pylori infection. Arch Dis Child 2000;83:429-34.

Kerr JR, Barson AJ, Burnie JP. Further information on an association between sudden infant death syndrome (SIDS) and Helicobacter pylori infection. Arch Dis Child (in press).

Malaty HM, Bousylman JE, Graham DY. Helicobacter pylori infection in day care centers (abstract). Gut 1996;39(suppl2):A85.

Mitz HS, Farber SS. Demonstration of Helicobacter pylori in tracheal secretions. J Am Osteopath Assoc 1993;93:87-91.

Mobley HL, Cortesia JM, Josenthal LE, Jones BD. Characterisation of urease from Campylobacter pylori. J Clin Microbiol 1988;26:831-6.

Nilsson HO, Taneera J, Castedal M, Glatz E, Olsson R, Wadstrom T. Identification of Helicobacter pylori and other Helicobacter species by PCR, hybridisation and partial DNA sequencing in human liver samples from patients with primary sclerosing cholangitis and primary biliary cirrhosis. J Clin Microbiol 2999;38:1072-6.

Pattison CP, Marshall BJ, Young TW, Vergara GG. Is Helicobacter pylori the missing link for sudden infant death syndrome? Gastroenterology 1997;112(4SS):A254.

Pattison CP, Smoot DT, Ashtorab H, Vergara GG, Young TW, Smith GP. Confirmation of Helicobacter pylori by PCR in sudden infant death syndrome. Gastroenterology 1998a;114(4):G3686.

Pattison CP, Marshall BJ, Scott LW, Herndon B, Willsie SK. Proposed link between Helicobacter pylori and sudden infant death syndrome (SIDS): possible pathogenic mechanisms in an animal model. I. Effects of intratracheal urease. Gastroenterology 1998b;114(4):G3689.