BREATH HYDROGEN AS A DIAGNOSTIC METHOD FOR HYPOLACTASIA

1155

BREATH HYDROGEN AS A DIAGNOSTIC

We have tested a simple method for the collection of exhaled H2 in the clinical situation.9 After an oral lactose load we compared the rise in breath H2 concentration with symptoms, maximum rise in bloodglucose, and, in patients requiring jejunal biopsy, with mucosal lactase activity.

METHOD FOR HYPOLACTASIA DAVID J. A. JENKINS ALVIN NEWMAN * LAURENCE M. BLENDIS

GEOFFREY METZ TIMOTHY J. PETERS

Department of Gastroenterology and Medical Research Council Gastroenterology Unit, Central Middlesex

Hospital,

and

Royal Postgraduate

Medical

Patients and Methods

School,

Twenty-five consecutive patients investigated in the

London

unit either for diarrhoea or abdominal symptoms of unknown cause were given 50 g. of lactose in 200 ml. of water by mouth. Any gastrointestinal symptoms such as borborygmi, excessive flatus, pain, discomfort, or diarrhoea were recorded. Samples of endexpiratory alveolar air from a single breath using a modified Haldane-Priestley tubewere collected every 30 minutes, and the H2 concentration was measured by gas chromatography. The results, expressed as change in H2 concentration at 120 minutes compared with fasting level, were compared with the maximum rise in bloodglucose during that period. Jejunal biopsy was done in patients with specific indications, part of the tissue being used for routine histology and part for assay of lactase activity. Disaccharidase activities were assayed independently by a modification 11 of the technique of Dahlqvist.12 Results were expressed as nmoles substrate hydrolysed per minute per mg. protein.13 In these cases the mucosal lactase activity was compared with the other variables being studied. The criterion for hypolactasia used in the study was a mucosal lactase activity of less than 100 milliunits per mg. protein. In the patients who did not have a biopsy, a maximum rise in blood-glucose of 20 mg. per 100 ml. or more was considered normal.

gastroenterology

(H2), collected by endexpiratory sampling, was measured in twenty-five patients with abdominal symptoms or diarrhœa after ingesting 50 g. of lactose. This was compared with established tests of hypolactasia. Fifteen patients with a blood-glucose rise of more Sum ary

Breath hydrogen

than 20 mg. per 100 ml. had less than 4 parts per million (p.p.m.) rise in breath H2 at 2 hours. In contrast, ten patients with blood-glucose rises of less than 20 mg. per 100 ml. had more than a 20 p.p.m. H2 rise (mean 85·8 p.p.m.±S.D. 44·3) at 2 hours. Similarly, two patients with normal jejunal lactase activity had no significant H2 production, whereas six patients with hypolactasia had more than a 20 p.p.m. rise in H2. Symptoms related to milk or lactose ingestion were found to be unreliable. End-expiratory sampling of breath H2 would seem to be a simple, non-invasive, and accurate method of diagnosing hypolactasia, which is also very acceptable to patients. This should make it a valuable tool both in diagnostic gastroenterology and in epidemiological surveys.

Results

Introduction SINCE lactose malabsorption is common in adultsi a rapid and simple method for its detection would be of great potential use in the investigation of patients with unexplained abdominal complaints such as diarrhoea, flatulence, or pain, and also more generally in studies of the incidence of lactose intolerance in different ethnic groups. At present the diagnosis depends on recording gastrointestinal symptoms and measuring the blood-glucose rise during a lactosetolerance test (L.T.T./.2 However, the mucosal lactase activity of jejunal-biopsy specimens remains the ultimate criterion. This must be so since symptoms after lactose ingestion can often be misleading and the changes in glucose levels in peripheral blood are influenced by other factors apart from the absorption

’I I

I II

B

I

Six patients, on direct questioning before the test, had the impression that milk was a cause or exacerbating feature of their symptoms. Four were shown to be intolerant to the lactose load, but two had completely normal results on testing (table i). During the L.T.T. fifteen of the twenty-five patients had a maximum blood-glucose rise of greater than 20 mg. per 100 ml., and they all had a 2-hour H2 rise of less than 4 parts per million (p.p.m.) (figure). They were also symptom-free, except one young woman being investigated for chronic flatulence who com-

<

I I I

I !’I

I,

I I I

process.3 Hydrogen (H2) is evolved when carbohydrate is fermented by bacterial Thus when malabsorbed lactose reaches the caecum it is fermented with production of H2. A proportion of the gas diffuses into the circulation to be carried to the lungs and exhaled.6 Indeed, in people diagnosed as lactose intolerant on the basis of a poor rise in blood-sugar, both the H2 per minute and its concentration in expired produced air increased after lactose ingestion.7.8 Unfortunately, equipment used for gas collection in this earlier work and the duration of collection-time may have prevented the methods from being generally accepted for clinical use. " Present address: New Mount Sinai Hospital, Toronto, Ontario, Canada.

Rise in Hz at

2hr.(pp.m.)

Maximum rise in blood-glucose compared with rise in endexpiratory H, concentration 2 hours after ingestion of lactose. Dotted lines show limits of normal (lower in glucose, upper

in H,).

1156 TABLE I-RESULTS FOR ALL

plained of her usual symptom of wind despite no H2 production, a blood-glucose rise of 37 mg. per 100 ml., and no other gastrointestinal symptoms. The remaining ten patients had a glucose rise of less than 20 mg. per 100 ml., and in all of these the 2-hour H2 rise was greater than 20 p.p.m. Eight of the ten had symptoms which varied from borborygmi (two) to bloating and wind (three), and in three cases bloating, wind, and diarrhoea were noted. One patient had no symptoms at all (despite a glucose rise of only 5 mg. per 100 ml. and H2 concentration of 27 p.p.m.) and another had a normal bowel action but no other symptoms during the test (with glucose rise 3 mg. per 100 ml. and H2 rise of 78 p.p.m.). All ten patients found to be lactose intolerant were put on a lactose-free diet. Two became symptomfree, their bowel habit returning to normal, and two others noted a decrease in symptoms. Of the eight patients who underwent jejunal biopsy, the six who had lactase levels below the normal range all had alveolar H2 concentrations greater than 20 p.p.m. and glucose rises less than 20 mg. per 100 ml. The remaining two patients with normal enzyme activity had glucose rises greater than 20 mg. per 100 ml. with no H2 production (table 11).

Discussion Patients with hypolactasia do not always give a history of milk intolerance 14 Therefore, a simple diagnostic test is required, since, in patients complaining of diarrhoea or flatulence, substitution of a lactose-free diet

render the patient symptom-free. patients intolerant to lactose develop Although symptoms during a standard SO g. L.T.T., many do not, despite evidence from blood-glucose changes or from breath H2 rise that part of the sugar load has been malabsorbed 15 On the other hand, some patients will note wind, flatulence, or borborygmi during a test when they have no evidence of hypolactasia. Thus, symptoms alone do not permit a diagnosis of lactose intolerance. can

most

25

PATIENTS

At present the maximum glucose rise after a 50 g. oral lactose load is used. However, there are many disadvantages in this method since changes in glucose concentration are dependent on gastric emptying and the rate of blood-glucose metabolism besides the mucosal absorptive function which one is testing, and it requires multiple blood-samples. The most conclusive diagnosis of hypolactasia is obtained by measuring lactase activity in jejunalbiopsy specimens. However, this technique requires intubation together with careful biochemical analyses. H2 is evolved and can be measured in expired air when even small amounts of fermentable substrate Its measurement come into contact with bacteria. involves neither venepuncture nor the swallowing of a tube and so it is more acceptable to patients. In addition, changes in H2 concentration are an effect of maldigestion, the condition which one is testing. Determination of lactose intolerance by breath analysis has been previously reported, samples being analysed either from total collections of expired air’ or after rebreathing for a fixed period of time in a closed circuit.5 Although the single end-expiratory sampling technique introduces greater potential inaccuracies, it produces a H2 concentration approaching the mixed-venous H2 level which is much greater than the concentration obtained from a total collection. Rebreathing in a closed circuit is the most accurate method of assessing total H2 production and produces high H2 concentrations, but the equipment needed and duration of collection are such that the technique has not been generally applicable to clinical situations. H2 evolution after lactose ingestion has not previously been compared with mucosal lactase levels. In this study hypolactasia is associated with a rise in H2 concentration at 2 hours of greater than 20 p.p.m. in every case while those patients with normal lactase levels have no significant H2 production. In addition to its potential diagnostic value in gastrointestinal disease, the detection of lactose

1157

intolerance is of great interest today in population

Although much surveys and paediatric medicine. research in many parts from been has work published 16-22 in remain world of the any map showlarge gaps of lactose tolerance and intoleradult ing the incidence ance 23

Clearly,

any test which eliminates

venepunctures would

simplify

multiple

the work in these

considerably. In all cases end-expiratory sampling has given a similar result to blood-glucose rise in the estimation of hypolactasia, and it could therefore be simply applied as a diagnostic test in gastrointestinal dissurveys

orders or in field surveys where the incidence of lactose intolerance is to be determined in a general

population. We thank Dr E. N. Rowlands, Dr J. J. Misiewicz, and Dr T. D. Kellock for their helpful advice and criticism. G. M. is in receipt of the T. K. Stubbins fellowship of the Royal College of Physicians. Requests for reprints should be addressed to L. M. B., Department of Gastroenterology, Central Middlesex Hospital,

London NWIO 7NS. REFERENCES

Cuatrecasas, P., Lockwood, D. H., Caldwell, J. R. Lancet, 1965, i, 14. 2. Bayless, T. M., Rosensweig, N. S., Christopher, N., Huang, S. S. Gastroenterology, 1968, 54, 475. 3. Newcomer, A. D., McGill, D. B. ibid. 1966, 50, 340. 4. Galloway, D. H., Mathews, R. D., Colasito, D. J. Nature, 1966, 1.

212, 1238. 5. Levitt, M. D. New Engl. J. Med. 1969, 281, 122. 6. Levitt, M. D., Ingelfinger, F. J. Ann. N.Y. Acad. Sci. 1968, 150, 75. 7. Calloway, D. H., Murphy, E. L., Bauer, D. Am. J. dig. Dis. 1969, 14, 811. 8. Levitt, M. D., Donaldson, R. M. J. Lab. clin. Med. 1970, 75, 937. 9. Metz, G. L., Gassull, M. A., Leeds, A. R., Blendis, L. M., Jenkins, D. J. A. Spring, 1975, meeting of British Society of Gastroenterology, abstracts. 10. Bond, J. H., Levitt, M. D. J. clin. Invest. 1972, 51, 1219. 11. Guilbault, G. G. Analyt. Chem. 1966, 38, 527. 12. Dahlqvist, A. Enzym. biol. clin. 1970, 2, 52. 13. Hiraoka, T., Glick, D. Analyt. Biochem. 1963, 5, 497. 14. Chenoweth, W. L. PH.D. dissertation, University of California,

Berkeley. 15. Isoski, M., Jussila, J., Sarna, S. Gastroenterology, 1972, 62, 28. 16. Kretchmer, N., Ransome-Kuti, O. Proc. Inst. Med. Chic. 1970,

28, 213. 17. Olatunbosun, D. A., Adadevoh, B. K. Am. J. dig. Dis. 1971, 16, 909. 18. Alzate, H., Gonzalez, H., Guzman, J. Am. J. clin. Nutr. 1969, 22, 122. 19. Gudmand-Hoyer, E., Jarnum, S. Acta med. scand. 1969, 186, 235. 20. Spanidou, E. P., Petrakis, N. L. Lancet, 1972, ii, 872. 21. Chung, M. H., McGill, D. B. Gastroenterology, 1968, 54, 225. 22. Bolin, T. D., Morrison, R. M., Steel, J., Davis, A. E. Med. J. Aust. 1970, i, 1289. 23. Simoons, F. J. Am. J. dig. Dis. 1973, 18, 595.

"

The process by which final demands upon the service determined is complex and certainly involves many non-medical considerations which remain only poorly understood. What emerges as certain is that demand tends

are

outstrip supply. Rationing, however, has never been explicitly organised but has hidden behind each doctor’s clinical freedom to act solely in the interests of his patient. Any conflict of interest between patients has been implicitly resolved by the doctor’s judgments as to their relative need

to

for care and attention. The clinical freedom to differ widely as to their conception of need has led to inconsistencies of treatment between patients and to the allocation, without challenge, of scarce resources to medical practices of no proven value."—MICHAEL H. CoopER, Rationing Health

Care;

p. 107.

London, 1975.

RUBELLA-VIRUS INFECTION IN JUVENILE RHEUMATOID ARTHRITIS Y. CHIBA J. L. DZIERBA

P. L. OGRA S. S. OGRA

J. K. HERD Departments of Pediatrics and Microbiology, School of Medicine, State University of New York at Buffalo, and Division of Virology, Children’s Hospital, Buffalo, N.Y.

Summary

14222, U.S.A.

Antibody

activity

against

mumps,

measles, polio, and rubella viruses was

determined in patients with juvenile rheumatoid arthritis (J.R.A.), rubella-vaccine associated arthritis, adult rheumatoid arthritis, other chronic systemic disorders (e.g., systemic lupus and dermatomyositis), and in a matched population of normal, non-rheumatoid (control) children. The antibody levels against mumps, measles, and poliovirus were similar in all patients. Rubella-antibody levels in rheumatoid arthritis and other systemic disorders were similar to those observed in controls. The mean rubella-antibody levels in rubella-vaccine arthritis were 4 times higher than in controls. The IgM and IgG rubellaantibody levels in J.R.A. were found to be 4-6 times higher when compared to titres observed in the controls. Highest antibody levels were seen in younger children with J.R.A. Detection of rubella-virus antigen was attempted by immunofluorescence in the sediment smears of synovial fluid of patients with J.R.A., adult rheumatoid arthritis, and other nonrheumatoid joint diseases. Specific staining for rubella virus antigen was observed in the synovial fluid of 33 % of patients with J.R.A. No antigen was detected in the synovial fluid from other patients. These observations suggest a possible role of rubella-virus infection in J.R.A.

Introduction AN association between acute arthritis and natural or vaccine-induced rubella virus infection has been amply demonstrated.1-6 Polyarthritis and polyarthralgia are frequent presenting manifestations of natural rubella in adult women.6 With the introduction of live attenuated vaccines, several investigators have defined the clinical and epidemiological features of arthritis associated with rubella-virus vaccines. In general, the joint manifestations of rubella-associated arthritis are transient: only rarely do joint symptoms persist for months after immunisation.8.7 The mechanism for the joint disease in rubella is poorly understood. The manifestations of arthritis have been related to the presence of infectious rubella virus in affected joints.7-9 There may be a positive relation between the development of arthritis, the appearance of rubella antibody in serum and joint fluid, and the presence of rubella virus in the affected joints Although rubella arthritis bears many clinical similarities to juvenile rheumatoid arthritis (j.R.A.), there is no evidence of a direct role of rubella infection in the aetiology or pathogenesis of chronic arthritis of childhood. We have investigated the serum-antibody response to naturally acquired or vaccine-induced viral infec-