Chest ACCP Member Benefits
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Guest Access | Sign In via User Name/Password
doi:10.1378/chest.06-1906
(Chest. 2007; 131:1166-1172)
© 2007 American College of Chest Physicians
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Thomson, R. M.
Right arrow Articles by Looke, D. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Thomson, R. M.
Right arrow Articles by Looke, D. F.

Gastroesophageal Reflux Disease, Acid Suppression, and Mycobacterium avium Complex Pulmonary Disease*

Rachel M. Thomson, MBBS; John G. Armstrong, MBBS, PhD and David F. Looke, MBBS

* From the Department of Thoracic Medicine (Dr. Thomson), The Prince Charles Hospital, Chermside; and the Departments of Respiratory Medicine (Dr. Armstrong) and Infectious Diseases (Dr. Looke), Princess Alexandra Hospital, South Brisbane, QLD, Australia.

Correspondence to: Rachel Thomson, MBBS, Department of Thoracic Medicine, The Prince Charles Hospital, Rode Rd, Chermside, QLD, Australia 4064; e-mail: R.Thomson{at}mailbox.uq.edu.au

Abstract

Background: Weekly symptoms of gastroesophageal reflux disease (GERD) occur in 20% of the population, and GERD has been implicated in the pathophysiology of many respiratory diseases. Microaspiration of contaminated water is a potential portal of entry for Mycobacterium avium complex (MAC) organisms into the respiratory tract, and acid-suppression therapy may enhance the survival of mycobacteria in the stomach. This study aimed to assess the prevalence of GERD, swallowing disorders, reflux symptoms, and acid-suppression therapy in patients with MAC lung disease (MAC positive [MAC+]), and to compare these patients to control subjects without MAC lung disease (MAC negative [MAC–]).

Methods: Clinical information was collected on 58 MAC+ patients and 58 age- and sex-matched MAC– patients who were asked to complete a DeMeester questionnaire of reflux symptoms and to identify any acid-suppressive medication consumed.

Results: A clinical diagnosis of GERD was documented in 23 of 52 MAC+ patients (44.2%), compared to 16 MAC– patients (27.6%) [p = 0.019]. MAC+ patients consumed significantly more histamine type 2 receptor antagonists and prokinetic agents, and MAC– patients consumed more antacids. The mean DeMeester questionnaire score (± SD) for MAC+ patients was 1.39 ± 1.8, and for MAC– patients was 0.88 ± 1.4. (p = 0.098). Aspiration was suspected in nine MAC+ patients (15.5%), compared to three MAC– patients (5.2%) [p = 0.032]. There was no association between GERD and radiologic presentation of MAC disease. Consolidation and nodules > 5 mm were more common in those receiving acid suppression than those who were not.

Conclusions: GERD, acid suppression, and clinically suspected aspiration are more common in patients with MAC lung disease than in similar patients without MAC disease.

Key Words: bronchiectasis • esophagus • gastroesophageal reflux • infection • mycobacterial

In HIV-seronegative individuals, factors affecting susceptibility to infection with Mycobacterium avium and Mycobacterium intracellulare and progression to disease are poorly understood. Patients are typically either middle-aged men with underlying chronic lung disease such as COPD, who have upper-lobe cavity formation and nodules of various sizes, or elderly patients with nodules, bronchiolitis, and bronchiectasis involving the middle lobe and lingula. These latter patients are more commonly women, nonsmokers, with no preexistent lung disease.12345

M avium complex (MAC) organisms are environmental, commonly found in natural waters, drinking water, and soils. They can be isolated from biofilms, aerosols, and dust.6 The state of Queensland lies in northeast Australia. M avium is capable of growth between 10°C and 45°C, typical temperatures for this area. Not surprisingly, therefore, exposure to MAC organisms in Queensland is common, as shown by results of avian Mantoux testing in children.78 The optimal pH for growth of most environmental mycobacteria is at acidic values, with little growth occurring at values of pH > 7.5. The optimal growth pH for M intracellulare is between 5 and 6.5.6

Symptoms of gastroesophageal reflux disease (GERD) are reported to occur weekly in approximately 20% of the population.91011 Both acid reflux and nonacid reflux have been implicated in the pathophysiology of chronic cough, asthma, idiopathic pulmonary fibrosis, and laryngeal dysfunction.1213 Aspiration of upper respiratory tract bacteria into the lungs is thought to occur quite frequently (in approximately 50% of normal hosts during sleep)14 and is generally of little consequence in the majority on individuals. Microaspiration of organisms probably occurs more frequently in the setting of altered sensorium, such as with alcohol or sedative drug use, and has been postulated as a cause of lower respiratory tract infections and chronic lung disease,1415 particularly if there is a large inoculum of bacteria, solid material, or acid. Deposition of relatively few mycobacteria can infect the host; estimates range from 2 to 3 bacteria, to as many as 50.16 Esophageal disorders, particularly achalasia, have been described in association with pulmonary disease due to rapidly growing mycobacteria.171819 A similar association for slow-growing mycobacteria has not been established. We postulated that if acid-suppression therapy enhances the survival of mycobacterial organisms in the stomach, then aspiration of gastric/oropharyngeal contents may be a potential portal of entry of these organisms into the respiratory tract.

The aim of this study was to assess the prevalence of a medical diagnosis of GERD, reflux symptoms, and acid-suppression therapy in patients with MAC lung disease (MAC positive [MAC+]), and to compare these patients with control subjects without MAC lung disease (MAC negative [MAC–]). We also aimed to assess the prevalence of reported swallowing disorders in these patients, who were at risk of aspiration.

Materials and Methods

A prospective cohort of 58 patients (cases) with MAC lung disease was identified by laboratory notification of isolates of MAC in Queensland, Australia between January 1 and December 31, 1999. The protocol (138/98) was approved by the Princess Alexandra Hospital Research Ethics Committee in accordance with the Australian National Health and Medical Research Council guidelines. Consent to participate was sought, and clinical features of the disease (including risk factors, symptoms, radiology, and treatment prescribed) were then provided by the treating doctor. Each patient was asked to complete a DeMeester questionnaire of reflux symptoms (Fig 1 ). A maximum score of 3 points was assigned for each of heartburn, regurgitation, and dysphagia. Patients were also asked to identify from a comprehensive list any acid-suppressing medication they consumed regularly. A clinical diagnosis of GERD by the treating physician was recorded; when possible, supportive gastroscopy or pH probe findings were obtained.


Figure 1
View larger version (37K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. DeMeester symptom score for GERD. *Mylanta (Pfizer Consumer HealthCare; West Ryde, NSW, Australia), Gaviscon (Reckitt Benckiser; West Ryde, NSW, Australia), and Quick-eze (Nestle; Blacktown; NSW, Australia).

 
Control patients (MAC–) were recruited from the Respiratory Outpatients Clinics. Each MAC+ case was matched to a MAC– control according to age (10-year groupings), gender, and smoking history (current smoker, ex-smoker, or nonsmoker). As other lung diseases can be associated with GERD, to minimize confounding, MAC– control subjects were matched according to the preexisting lung disease of the MAC+ cases (eg, COPD, preexisting bronchiectasis). If patients with mycobacterial disease had no preexistent lung disease, then they were matched with control subjects also without lung disease, or in some patients with mild COPD or asthma, but with similar respiratory symptoms such as chronic cough. When the underlying lung disease was COPD or bronchiectasis, the control subjects were required to have a minimum of three negative sputa for mycobacteria within the preceding 12 months and no history of mycobacterial disease. Each MAC– control subject was also asked to complete a DeMeester symptom questionnaire and to record any GERD medication consumed. Gastroscopy, pH probe, and radiographic contrast swallow findings were also recorded. Due to the difficulty in finding matched control subjects, they were not matched for factors such as steroid use, alcohol consumption, or diabetes; however, these were recorded and included in the statistical analysis. While obesity can be associated with an increased risk of GERD, the cases and controls were not matched according to body mass index. As most patients with MAC lung disease are thin, this would likely bias toward the null, rather than confound the findings.

MAC+ patients were classified as having cavitary disease (any cavities present, including those associated with bronchiectasis) or noncavitary disease (predominantly nodular bronchiectasis). The radiographic distribution of disease was divided into upper lobe, middle lobe (right middle lobe and lingula), and lower lobe according to the predominant diseased areas. Patients receiving medication for GERD were grouped into antacids, histamine type 2 receptor antagonists (H2RAs), proton-pump inhibitors, and prokinetic agents.

Data were analyzed using statistical software (SPSS v12.0 for Windows 2003; SPSS; Chicago, IL). Tests of association were performed using Fisher exact test for {chi}2 two-by-two tables. DeMeester scores were compared using the Mann Whitney U test. DeMeester scores were also compared using a univariate analysis of variance that included both matched factors and potential confounders. Logistic regression analysis including matched factors and potential confounders was used to compare the prevalence of GERD between MAC+ and MAC– subjects.

Results

The details of MAC+ cases and MAC– controls are shown in Table 1 . Radiologic details of those with disease (MAC+) are shown in Table 2 and bacteriology in Table 3 .


View this table:
[in this window]
[in a new window]

  		Table 1. Characteristics of MAC+ and MAC– Patients*

 

View this table:
[in this window]
[in a new window]

  		Table 2. Radiologic Findings of MAC+ Patients

 

View this table:
[in this window]
[in a new window]

  		Table 3. Bacteriology

 
Clinical information about reflux disease was returned for 52 of 58 MAC+ patients. Twenty-three MAC+ patients (44.2%) had a diagnosis of GERD recorded by their treating doctor, compared to 16 MAC– patients (27.6%) [p = 0.019]. Confirmatory evidence was provided in 10 MAC+ patients: 6 patients had positive findings on gastroscopy, and 4 patients had radiologic evidence on Barium swallow. Ten MAC– patients had undergone gastroscopy, and 1 patient had a pH probe.

There were significant differences between MAC+ and MAC– patients in the use of acid suppression, as shown in Table 4 . MAC+ patients consumed significantly more H2RAs and prokinetic agents, and MAC– patients consumed more antacids. A logistic regression model that included both matched factors and potential confounders (alcohol intake, smoking history, underlying lung disease, and steroid use) confirmed a significant association between GERD and MAC disease (p = 0.008).


View this table:
[in this window]
[in a new window]

  		Table 4. Prevalence of GERD and Consumption of Acid-Suppressive Medication in Cases (MAC+) and Controls (MAC–)*

 
DeMeester questionnaire scores are shown in Table 5 . The mean score (± SD) for MAC+ patients was 1.39 ± 1.8, and for MAC– patients was 0.88 ± 1.4. As the scores were not normally distributed, they were compared using the Mann Whitney U test, which failed to show a statistically significant difference between the two groups (p = 0.098). Potential differences between MAC+ and MAC– groups were tested using an analysis of variance that included matched factors and other potential confounding factors (steroids, smoking history, underlying lung disease). There was a difference in DeMeester scores across both MAC+ and MAC– for steroid use with significant differences between those not receiving steroids and those receiving continuous oral steroids for > 6 months (p = 0.005). This difference was greater for MAC+ (p = 0.007) than MAC– (p = 0.068).


View this table:
[in this window]
[in a new window]

  		Table 5. DeMeester Questionnaire Scores for MAC+ vs MAC– Patients*

 
Nine MAC+ patients (15.5%) were thought to aspirate for other reasons. A summary of these patients is shown in Table 6 . All of these patients had bronchiectasis and nodules; three patients also had cavities. This compared to three MAC– patients in whom aspiration was suspected (p = 0.032).


View this table:
[in this window]
[in a new window]

  		Table 6. MAC+ Patients in Whom Aspiration Was Suspected*

 
There was no association between a medical diagnosis of GERD and radiologic presentation of MAC disease according to the groups outlined in Table 2, nor when grouped as cavitary vs noncavitary nor predominant distribution of disease (ie, upper/middle/lower lobes). More patients with consolidation on radiography consumed acid-suppressive medication regularly than those without consolidation (p = 0.018). This was mainly due to the relationship between H2RA use and consolidation. Seven of 17 patients (41.2%) with consolidation on radiography were receiving H2RAs, compared to 2 of 30 patients (6.7%) without consolidation (p = 0.007). Six of 11 patients (54.5%) with nodules > 5 mm consumed some form of acid suppression, compared to 8 of 36 patients (22.2%) without nodules > 5 mm (p = 0.06). There was no correction made for multiple comparisons.

Discussion

GERD is a common problem in patients with a number of respiratory diseases. The mechanics of coughing and the related reduction in lower esophageal sphincter tone encourage the reflux of gastric contents into the lower esophagus. The lower pH in the distal esophagus has been shown to cause further cough. Hence, proving a connection between GERD and MAC lung disease specifically was likely to be difficult.

Patients in this cohort with pulmonary MAC disease have a higher incidence of GERD than is reported in general population surveys and when compared to a matched cohort without MAC disease. The consumption of regular acid-suppressive medication by MAC+ patients was greater than MAC– control subjects. While DeMeester questionnaire scores of gastroesophageal reflux symptoms also appeared to be greater in MAC+ than MAC–, this failed to reach statistical significance and was accounted for by higher oral steroid use in those with higher scores. The difference however may also have been affected by the greater use of acid suppression in the MAC+ group. The consumption of acid-suppressive medication is likely to have underestimated the DeMeester questionnaire scores in these patients, although it did not abolish reflux symptoms in all patients. While the frequency of antacid use was not formally measured in either group, in the majority of patients it was not on a daily basis, suggesting that the symptoms of GERD were in fact less frequent or severe than those who required more potent and regular acid suppression.

The finding of more patients receiving acid suppression among those with consolidation and nodules has not been documented before. Consolidation is not specific for mycobacterial disease, and it is possible that other aspirated pathogens were responsible for coinfection is these cases. Acid suppression may be a marker for more severe GERD and hence greater risk of microaspiration. One could not exclude an effect of acid suppression on the nature of the aspirated refluxate that may have subsequently affected the inflammatory reaction to mycobacteria in the lung. However, as no correction was made for multiple comparisons, it is possible this finding occurred by chance.

A major limitation in this observational study is how reflux was defined. The recording of a medical diagnosis of reflux and the prescription of acid-suppressing medication imply that the patients were symptomatic. However, GERD can be silent. Gastroscopy and barium contrast studies both have limitations in diagnosing reflux disease, and as a result few patients had undergone these procedures. The 24-h pH probe has been considered the "gold standard," but even with this procedure, nonacid reflux events are not recognized, the height of the reflux above the gastroesophageal junction is poorly delineated, and the physical nature of the refluxed material is not distinguished. Given the invasive nature of this investigation, in the routine management of patients the pH probe is unlikely to change management unless symptoms persist on maximal medical therapy. Therefore, as this was an observational study (not interventional), few patients had undergone pH probe monitoring. Given also that this was a cross-sectional study, the timing of the diagnosis of GERD relative to the diagnosis of Mycobacterial disease or institution of treatment was not clear. The duration of GERD symptoms was not measured by the DeMeester questionnaire but would have been useful information. It is possible that some patients had reflux symptoms as a side effect of coughing or drugs used in treatment. However, in MAC lung disease there is often a long presymptomatic phase of disease; given the problems with recall bias and the already mentioned problems with establishing a diagnosis of reflux and the presence of silent reflux, information regarding the timing of diagnosis may not have been particularly helpful.

Interestingly, 9 of 58 patients (15.5%) with pulmonary MAC disease were thought to aspirate for other reasons. One must then ask, does acid suppression enhance the survival of organisms so that with repeated reflux episodes more are likely to be aspirated, or is the chance of infection related to the frequency and size of aspiration episodes of contaminated food/liquid?

In vitro work20 would support the latter theory. M avium has been found to be naturally tolerant to pH < 3, and preadaptation, under conditions similar to the conditions where M avium is found in the environment, resulted in increased acid resistance. One would expect then that acid suppression would have little additional effect on the survival of M avium in the stomach. A Japanese study21 found that omeprazole exhibited no significant effect on mycobacterial growth in human alveolar macrophages or on the mycobactericidal activity of clarithromycin against them.

One might conclude then that acid suppression would have little effect on the risk of acquiring MAC, as the organisms are likely to survive gastric acid anyway. However, the use of acid suppression may be a surrogate marker for more severe reflux disease and hence indicate a greater chance of microaspiration. This theory is supported by the observation of greater use of acid suppression in MAC+ patients than MAC– control subjects, particularly prokinetic agents. The finding of a significant number of patients with nodules and bronchiectasis thought to aspirate as a result of oropharyngeal dysfunction also raises the possibility that microaspiration/macroaspiration may be a factor involved in the etiology of nodular bronchiectasis. This would make accurate assessment of reflux more important in these patients, as nonacid, liquid reflux reaching a level above the gastroesophageal junction to make aspiration likely would be of interest. Balaji and colleagues22 described impedance technology, which has the capacity to detect all types of reflux (acid, nonacid, liquid, mixed, and air). In their evaluation of this technology in normal individuals, they showed that over half of reflux events are not detected by pH studies and liquid reflux reached to the mid and proximal esophagus 69% of the time. With the acid resistance of slow-growing mycobacteria, this technique would seem appropriate in evaluating MAC patients for reflux as a means of acquisition of mycobacterial infection. Patients presenting with MAC lung disease should be assessed for GERD and swallowing disorders, but larger prospective studies more formally evaluating these patients for swallowing disorders and GERD are warranted.

Acknowledgements

The authors thank Dr. Chris Coulter and David Dawson from the QLD Mycobacterial Reference Laboratory, Dr. Jenny Robson from Sullivan and Nicolaides Pathology, and Dr. Ross LeFeuvre and Robyn Fraser from Queensland Medical Laboratory for their assistance in providing reports of all mycobacterial isolates during 1999, and subsequent isolates from the patients in this cohort. Clinical support was provided by Dr. Tom Konstantinos and the Queensland TB Control Unit, Brisbane, QLD, Australia, for which we are also grateful.

Footnotes

Abbreviations: GERD = gastroesophageal reflux disease; H2RA = histamine type 2 receptor antagonist; MAC = Mycobacterium avium complex; MAC+ = Mycobacterium avium complex positive; MAC– = Mycobacterium avium complex negative

This research was supported by a Research Scholarship from the Princess Alexandra Hospital, South Brisbane, QLD, Australia, and was conducted through the Department of Thoracic Medicine of the Princess Alexandra Hospital.

The authors have no conflicts of interest to declare.

Received for publication August 3, 2006. Accepted for publication December 13, 2006.

References

  1. Reich, JR (1995) Nontuberculous mycobacteria: an underappreciated cause of geriatric lung disease. Am J Respir Crit Care Med, 151,922-923[ISI][Medline]
  2. Shih, JYHP, Lee, LN, Wang, HC, et al Nontuberculous mycobacteria isolates: clinical significance and disease spectrum. J Formos Med Assoc 1997;96,621-627[ISI][Medline]
  3. Wallace, RJ, Jr, Zhang, Y, Brown, BA, et al Polyclonal Mycobacterium avium complex infections in patients with nodular bronchiectasis. Am J Respir Crit Care Med 1998;158,1235-1244[Abstract/Free Full Text]
  4. Debrunner, M, Salfinger, M, Brandli, O, et al Epidemiology and clinical significance of nontuberculous mycobacteria in patients negative for human immunodeficiency virus in Switzerland. Clin Infect Dis 1992;15,330-345[ISI][Medline]
  5. Pulmonary disease caused by Mycobacterium avium-intracellulare in HIV-negative patients: five-year follow-up of patients receiving standardised treatment. Int J Tuberc Lung Dis 2002;6,628-634[ISI][Medline]
  6. Falkinham, J, III Nontuberculous mycobacteria in the environment. Clin Chest Med 2002;23,529-552[CrossRef][ISI][Medline]
  7. Abrahams, EWHR Sensitivity to avian and human PPD in Brisbane school children. Tubercle 1967;48,79-94[Medline]
  8. Abrahams EW HR.. Studies with the purified protein derivative of human and avian tuberculins in South Queensland. Tubercle 1968;49,192-209[CrossRef][Medline]
  9. Isolauri, J, Laippala, P Prevalence of symptoms suggestive of gastro-esophageal reflux disease in an adult population. Ann Med 1995;27,67-70[ISI][Medline]
  10. Corder, AP, Jones, RH, Sadler, GH, et al Heartburn, oesophagitis and Barrett’s oesophagus in self-medicating patients in general practice. Br J Clin Pract 1996;50,245-248[ISI][Medline]
  11. Locke, GR, III, Talley, NJ, Fett, SL, et al Prevalence and clinical spectrum of gastroesophageal reflux: a population based study in Ohlmsted County, Minnesota. Gastroenterology 1997;112,1448-1156[CrossRef][ISI][Medline]
  12. Field, SK, Field, TS, Cowie, RL Extraesophageal manifestations of gastroesophageal reflux. Minerva Gastroenterol Dietol 2001;47,137-150[Medline]
  13. Raghu, G, Freudenberger, TD, Yang, S, et al High prevalence of abnormal acid gastro-oesophageal reflux in idiopathic pulmonary fibrosis. Eur Respir J 2006;27,136-142[Abstract/Free Full Text]
  14. Huxley, EJ, Viroslav, J, Gray, WR, et al Pharyngeal aspiration in normal adults and patients with depressed consciousness. Am J Med 1978;64,564-568[CrossRef][ISI][Medline]
  15. Wynne, JW, Modell, JH Respiratory aspiration of stomach contents. Ann Intern Med 1977;87,466-474[ISI][Medline]
  16. Mason, CM, Weinacker, AB, Shellito, JE Host defense networks in the lung: an overview. Neidermann, MS Sarosi, GA Glassroth, J eds. Respiratory infections. 2001 Lippincott, Williams & Wilkins. Philadelphia, PA:
  17. Hadjiliadis, D, Adlakha, A, Prakash, UB Rapidly growing mycobacterial lung infection in association with esophageal disorders. Mayo Clin Proc 1999;74,45-51[ISI][Medline]
  18. Hutchins, GM, Boitnott, JK Atypical mycobacterial infection complicating mineral oil pneumonia. JAMA 1978;240,539-541[Abstract]
  19. Karsell, PR Achalasia, aspiration, and atypical mycobacteria. Mayo Clin Proc 1993;68,1025-1026[ISI][Medline]
  20. Bodmer, T, Miltner, E, Bermudez, LE Mycobacterium avium resists exposure to the acidic conditions of the stomach. FEMS Microbiol Lett 2000;182,45-49[CrossRef][ISI][Medline]
  21. Suzuki, K, Tsuyuguchi, K, Matsumoto, H, et al Effect of proton pump inhibitor alone or in combination with clarithromycin on mycobacterial growth in human alveolar macrophages. FEMS Microbiol Lett 2000;182,69-72[CrossRef][ISI][Medline]
  22. Balaji, NS, Blom, D, DeMeester, TR, et al Redefining gastroesophageal reflux (GER). Surg Endosc 2003;17,1380-1385[CrossRef][ISI][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Thomson, R. M.
Right arrow Articles by Looke, D. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Thomson, R. M.
Right arrow Articles by Looke, D. F.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS