9

OSTEOMALACIA AFTER SMALL-INTESTINAL RESECTION L. W. L. HORTON J. R. TIGHE B. CREAMER

JULIET E. COMPSTON A. B. AYERS

Departments of Gastroenterology, Surgical Pathology, and

Radiology, St.

Thomas’ Hospital, London SE1

Histological examination of bone from 25 patients with small-intestinal resection showed that 9 (36%) had osteomalacia, which was severe in 5 and mild in 4. The serum-alkaline-phosphatase concentration was raised in all patients with severe osteomalacia, but serum calcium, phosphate, and alkaline-phosphatase concentrations were normal in the 4 patients with mild disease, 2 of whom had symptoms. Osteomalacia was diagnosed radiologically in only 3 paSummary

tients.

Osteomalacia appears to be commoner in patients with small-intestinal resection than has previously been thought, and bone biopsy is essential if all cases are to be detected. Although high-dose parenteral vitamin-D therapy is usually effective in the treatment of osteomalacia after small-intestinal resection, our findings showed that oral vitamin-D metabolites and their analogues may also be effective. This has important

practical advantages. Introduction

OSTEOMALACIA may occur in patients with smallintestinal disease or resection, but its prevalence in a large series has not been previously reported. Factors contributing to the development of osteomalacia in these patients include malabsorption of dietary vitamin-Dl and calcium, and of 25-hydroxyvitamin D (25-OHD) undergoing enterohepatic circulation;2 reduced exposure to ultraviolet light; a low dietary vitamin-D intake; and the ingestion of drugs altering the metabolism3 or intestinal absorption4 of vitamin D. Winter plasma levels of 25-OHD, the major circulating metabolite of vitamin D, are significantly reduced in patients with small-intestinal resection compared with normal controls,S indicating

that some degree of vitamin-D deficiency is common in these patients. Histological osteomalacia may exist without clinical, biochemical, or radiological abnormalities,6.7 and therefore bone biopsy is essential if all cases are to be detected. We report on the prevalence of histological osteomalacia and its relationship to clinical, biochemical, and radiological variables in 25 patients with smallintestinal resection. We also describe the response of 4 patients with severe osteomalacia to some vitamin-D

preparations given orally or parenterally. Patients and Methods Patients and Controls 25

patients (17

women, 8

men), aged

22-78 yr

(mean 42),

studied. The reason for resection was Crohn’s disease in 22, radiation enteritis in 1, volvulus in 1, and a congenital abnormality in 1. 20 patients had undergone ileal resection, 2 were

and 3 both ileal and jejunal. The length of small intestine resected (measured in 16 patients) ranged from 30 to 194 cm (mean 96-3) and the length of intestine remaining after resection (measured in 19 patients) ranged from 30 to 390 cm (mean 201). 10 patients had recurrent small-intestinal Crohn’s disease at the time of the study, demonstrated radiologically or at laparotomy. The mean number of years since the first resection was 6.4 (range 1-14). Apart from 6 patients who were taking prednisone, no patient was taking any drug known to induce hepatic microsomal enzymes, nor was any patient receiving calcium or vitamin-D supplements. Control bone biopsies were obtained at necropsy from 7 women and 19 men, aged 29-75 yr (mean 53), who had died suddenly and who had no previous history of immobilisation or metabolic bone disease.

jejunal,

Bone

Histology

. Full-thickness iliac-crest biopsies were obtained 2 cm below and behind the anterior superior iliac spine,8 and 8 µm undecalcified sections were stained by the Von Kossa technique and 1% toluidine blue. Calcification fronts were demonstrated by fluorescence microscopy of 10 m unstained sections after oral administration of 900 mg of demethylchlortetracycline 48 h before the biopsy. Histological quantitation was made with a Zeiss 25-point eye-piece graticule. Histological osteomalacia was diagnosed when there was an increased surface-area and volume of osteoid associated with a reduction in the percentage of osteoid seams bearing a calcification front.

Biochemistry I.A.R.C. Intestinal Microecology Group. Lancet, 1977, ii, 207. Burkitt, D. P., Walker, A. R. P., Painter, N. S. ibid. 1972, ii, 1408. Hoppert, C. A., Clark, A. J. J. Am. dietet. Assoc. 1945, 21, 157. Williams, R. D., Olmsted, W. H. J. Nutr. 1936, 11, 433. Cummings, J. H., Jenkins, D. J. A., Wiggins, H. S. Gut, 1976, 17, 210. Paul, A. A., Southgate, D. A. T. McCance and Widdowson’s The Composition of Foods. H.M. Stationery Office (in the press). 7. Southgate, D. A. T. J. Sci. Fdn. Agric. 1969, 20, 331. 8. Kirwan, W. O., Smith, A. N., McConnell, A. A., Mitchell, W. D., Eastwood, M. A. Br. med. J. 1974, iv, 187. 9. Hudson, G. J., John, P. M. V., Bailey, B. S., Southgate, D. A. T. J. Sci. Fd 1 2. 3. 4. 5. 6.

10.

Serum TABLE

calcium, phosphate, and alkaline-phosphatase

I-QUANTITATIVE

A. F. J. hum. Nutr. 1976, 30, 303. 11. Cummings, J. H. Unpublished. 12. Jelaca, S. L., Hlynka, I. Cereal Chem. 1971, 48, 211. 13. van Soest, P. J. Fedn. Proc. 1973, 32, 1804. 14. Eastwood, M. A., Girdwood, R. H. Lancet, 1968, ii, 1170. 15. Southgate, D. A. T., Durnin, J. V. G. A. Br. J. Nutr. 1970, 24, 517. 16. Dawson, A. M., Holdsworth, C. D., Webb, J. Proc. Soc. exp. Biol. Med. 1964, 117, 97. 17. McNeil, N. I., Cummings, J. H., James, W. P. T. Gut, 1977, 18, A425. 18. Jenkins, D. J. A., Leeds, A. R., Newton, C., Cummings, J. H. Lancet, 1975,  , 1116. 19. Jenkins, D. J. A., Hill, M. J., Cummings, J. H. Am. J. clin. Nutr. 1975, 28, 1408.

BONE HISTOLOGY IN PATIENTS AND CONTROLS

Agric. 1976, 27, 681. Southgate, D. A. T., Bailey, B., Collinson, E., Walker,

Results given as meanis.D.

con-

10 centrations

were

measured

on

the S.M.A.

Radiology

12/60 ’AutoAna-

lyzer’.

8 of the 9 patients with histological osteomalacia had metabolic bone surveys, but only 3 (table 11) showed definite osteomalacia-i.e., Looser zones or non-healing pathological fractures. In 5 patients, generalised bone

Radiology were carried out in 16 patients. In remaining 9 patients only a limited radiological assessment was possible, based mainly on films of the abdomen and chest.

Metabolic bone surveys

the

demineralisation (osteopenia) was seen, 1 case being associated with secondary hyperparathyroidism. 1 patient with severe histological osteomalacia had no radiological

Results

abnormalities. Metabolic bone surveys were done in 8 non-osteomalacic patients, and showed osteopenia in 1, osteoporosis in 2, and no abnormality in 5.

Bone Histology 9 patients (36%) had histological osteomalacia, which was severe in 5 and mild in 4. No significant differences in the

Comparison of Osteomalacic with Non-osteomalacic

of osteoid or calcification fronts were found between non-osteomalacic patients and controls amounts

Patients

(table i).

There

statistically significant differences groups in age, sex ratio, number of resection, length of intestine resected, or

were

between the

no

two

Clinical Symptoms (Table II)

years since

Bone pains were present in 3 patients with severe, and in 2 with mild, histological osteomalacia, and responded to vitamin-D therapy in all cases. Proximal myopathy was present in 1 patient with severe osteomalacia.

left behind. Recurrent Crohn’s disease was presin 44% of patients with osteomalacia and 37.5% of those without; 5 of the 6 patients taking prednisone were non-osteomalacic. The mean dietary vitamin-D intake was less than the recommended daily intake of 2-3 ug in both groups but was lower in the patients with osteomalacia (1.2 vs 1.88 tg).

length ent

Biochemistry (Table II) Serum-alkaline-phosphatase

was elevated in all patients with severe osteomalacia but was normal in the 4 with mild bone disease. Serum-calcium, corrected for serum-albumin,9 was low in only 2 patients with severe osteomalacia, and serum-phosphate was normal in all 9 patients with osteomalacia. Serum-calcium and serumphosphate were normal in all patients with normal bone histology, but serum-alkaline-phosphatase was raised in 2 (due to an elevation of the liver isoenzyme). TABLE

I!——CLINICAL, BIOCHEMICAL, I

*Full metabolic bone survey

not

Response to Treatment (Table III) 2 patients with severe osteomalacia showed a complete biochemical and histological response after treatment for 1 year with high doses of parenteral vitamin-D2. The third patient, who has been described in detail elsewhere,1O responded only partially to 2 years of parenteral vitamin-D2 therapy, but bone histology and

AND RADIOLOGICAL DATA IN THE NINE PATIENTS WITH OSTEOMALACIA I

I

I

I

done.

t2HPTD: secondary hyperparathyroidism. TABLE III-HISTOLOGICAL RESPONSE TO VITAMIN-D THERAPY IN FOUR OF THE PATIENTS WITH SEVERE OSTEOMALACIA

*i.m.: intramuscular. p.o.: oral.

tDose: A-120 000 units/week: B-60 000 units/week; C-Vit D2 120 000 units/week i.m. and 1 «OHD3 2 p.g/day; E-50 p.g/day.

11

biochemistry returned to normal after 6 months’ treatment with oral 1cx-hydroxyvitamin D3 (lotOHD3), a synthetic analogue of the active meabolite of vitamin D, 1,25-dihydroxyvitamin D3 (1,25-[OH]2 D3). The fourth patient with severe osteomalacia showed a dramatic histological response to oral 25-OHD3 after 6 months; accompanied by correction of hypocalcsemia and return of the serum-alkaline-phosphatase concentration tc normal. Discussion was surprisingly common in osteomalacia Histological this group of patients with small-intestinal resection, being present in approximately a third. The greater likelihood of patients with biochemical or radiological evidence of osteomalacia being biopsied introduced some element of selection, but the majority of patients biopsied had no such evidence of osteomalacia, and this series does not represent a highly selected group of

patients. This study clearly shows that bone biopsy is the only certain method of diagnosing osteomalacia; we found bone biopsy to be a safe, well-tolerated procedure which can be done on outpatients and which can be repeated several times. The clinical importance of mild histological osteomalacia and its natural history have not yet been established. It may not progress to severe disease, since improvement in general health, with increased endogenous vitamin-D synthesis during the summer," and increased vitamin-D and calcium intake could produce remission of the bone disease. However, 2 of our patients with mild histological osteomalacia complained spontaneously of bone pains, which responded to vitamin-D therapy, indicating that, in them, the histological findings were clinically significant. Severe osteomalacia can be a crippling and painful disease, and its prevention by early diagnosis and treatment is obviously desirable. On the basis of our findings we would recommend that ’all patients with small-intestinal resection who have any clinical, radiological, or biochemical abnormality suggestive of osteomalacia should undergo bone biopsy, and that the response of those patients with osteomalacia should be assessed by another histological examination of the bone after 6 months to a year of treatment. A raised serum-alkaline-phosphatase level was a constant finding in patients with severe histological osteomalacia, but was absent in all patients with mild disease; moreover, in patients with Crohn’s disease, abnormal liver-function may produce elevated serum-alkalinephosphatase concentrations, and reliable separation of the isoenzymes is not routinely available at present. Serum calcium and phosphate concentrations were not useful in diagnosing osteomalacia in this series, and although plasma 25-OHD levels give a reasonable indication of vitamin-D status, we have not found they correlate closely with bone histology; this has been confirmed by Driscoll et al.12 Radiology, although diagnostic in 3 cases, 1 of which showed only mild histological osteomalacia, was otherwise unhelpful. It is interesting that osteopenia was diagnosed in 6 patients, 5 of whom had histological osteomalacia, and we believe that with increased experience of correlating radiological and histological abnormalities of bone the diagnostic yield of metabolic bone surveys may be considerably improved by careful examination of the bone trabecular pattern and corticomedullary differentiation.

The majority of patients studied had Crohn’s disease, and in most of them small-intestinal disease had been present for some time before resection. Since patients were not usually biopsied at the time of their first resection, the possibility of osteomalacia being present then cannot be excluded. Disease affecting a length of small intestine may produce the same functional effect as resection, and resection per se may not have been the major factor in the development of osteomalacia in all

patients. study does not reveal factors which can be used to predict reliably which patients will develop osteomalacia, and a multifactorial setiology seems more likely than a single cause. Bile acids are essential for the absorption of vitamin D in man,13 and interruption of their enterohepatic circulation by ileal disease or resection may reduce absorption of dietary vitamin D and of 25-OHD undergoing enterohepatic circulation. Faecal fat excretion was not routinely measured in our patients, but the 2 patients with severe and 1 with mild osteomalacia who had short ileal resections (60, 60, and 30 cm) and no recurrent disease are unlikely to have had significant fat malabsorption.14 In all 9 patients with osteomalacia, at least one other factor was identified in addition our

Our

resection which could have contributed towards the development of bone disease. 2 patients had been virtually housebound for some years, and, in them, endogenous vitamin-D synthesis, normally the major source of circulating 23-OHD,"’ must have been minimal. Dietary vitamin-D intake was well below the recommended daily intake in 7 of the 9 patients with osteomalacia, and this may have been a contributory factor where borderline vitamin-D deficiency already existed. However, dietary vitamin D is an unimportant source of circulating 25-OHD under normal circumstances and many patients without osteomalacia had comparably low intakes. 3 patients developed osteomalacia after a period of rapid bone growth (which had been delayed during adolescence by their inflammatory bowel disease) when vitamin-D requirements are known to be increased, and this may represent a time when there is a special risk of developing osteorpalacia. Cholestyramine may have been a contributory factor in one patient with severe osteomalacia, since cholestyramine is known to reduce intestinal absorption of vitamin D and may also increase loss of 25-OHD from the enterohepatic circulation.l’ Finally, although the length of small intestine left behind after resection did not appear to be an important factor in the development of osteomalacia, 1 patient left with only 30 cm of jejunum had severe fat malabsorption and vitamin-A deficiency. to

large

doses of parenteral vitamin D normally osteomalacia associated with small-intesgiven tinal disease was effective in 2 patients (patients A and B) but it was only partially effective in another patient (patient C). Most patients prefer to take drugs orally, and, even in the presence of malabsorption, therapeutic blood-levels may be achieved provided the dose is sufficient. Hence 2 patients (patients C and E) responded rapidly to oral therapy despite severe fat malabsorption in 1 and continued cholestyramine therapy in the other. The use of vitamin-D metabolites and their analogues in the treatment of osteomalacia has not yet been fully evaluated but offers possible advantages. Their therapeutic dose-range is much narrower, so that the correct dose The

to treat

12 for each patient can be established more rapidly than with the parent vitamin, and their onset of action is more rapid. In addition, the short biological half-time of 1cxOHD) and 1,25-OHD3 (3-5 days)l8 enables hypercalcaemia associated with toxicity to be reversed much more rapidly than with the parent vitamin.19 25-OHD3, however, has a longer biological half-time (15-30 days)2O and from this point of view treatment with 1cxOHD3 or 1,25(OH)2D3 is preferable. We thank J. Sainsbury Ltd. for financial support, Mr Adrian Webb for technical assistance, Dr Norman Eve, Roussel Laboratories, for supplying 25-OHD,, and Mr Nigel Underwood, Leo Laboratories Ltd., for supplying synthetic 1 xOHD3.

for

Requests reprints should be addressed to J. E. C., Gastrointestinal Laboratory, Rayne Institute, St. Thomas’ Hospital, London, SE1. REFERENCES

1. 2.

G. R.,

Lewis, B., Booth, C. C. J. clin. Invest. 1966, 45, 94. Thompson, Arnaud, S. B., Goldsmith, R. S., Lambert, P. W., Go, V. L. W. Proc. Soc. exp. Biol. Med. 1975, 149, 570. 3. Dent, C. E., Richens, A., Rowe, D. J. F., Stamp, T. C. B. Br. med. J. 1970, iv, 69.

4. Thompson, W. G., Thompson, G. R. Gut, 1969, 10, 717. 5. Compston, J. E., Creamer, B. ibid. 1977, 18, 171. 6. Chalmers, J., Conacher, W. D. H., Gardner, D. L., Scott, P.

J. J. Bone Jt Surg. 1967, 49B, 403. 7. Melvin, K. E. W., Hepner, G. W., Bordier, P. J., Neale, G., Joplin, G. F. Q. Jl Med. 1970, 153, 83. 8. Bordier, P. J., Matrajt, H., Hioco, D. Path. Biol. 1964, 12, 1238. 9. Berry, E. M., Gupta, M. M., Turner, S. J., Burns, R. R. Br. med. J. 1973, iv, 640.

Compston, J. E., Horton, L. W. L., Tighe, J. R. ibid. 1977, ii, 612. Stamp, T. C. B., Round, J. M. Nature, 1974, 247, 563. Driscoll, R., Meredith, S., Wagonfeld, J., Rosenberg, I. Gastroenterology, 1977, 72, A-28/1051. 13. Blomstrand, R., Forsgren, L. Acta chem. scand. 1967, 21, 1662. 14. Hofmann, A. F., Poley, J. R. Gastroenterology, 1972, 62, 918. 15. Haddad, J. G., Hahn, T. J. Nature, 1973, 244, 515. 16. Preece, M. A., Tomlinson, S., Ribot, C. A., Pietrek, J., Korn, H. T., Davies, D. M., Ford, J. A., Dunnigan, M. G., O’Riordan, J. L. H. Q. Jl Med.

10. 11. 12.

1975, 176, 575. Compston, J. E., Thompson, R. P. H. Lancet, 1977, i, 721. Mawer, E. B., Backhouse, J., Davies, M., Hill, L. F., Taylor, C. M. ibid. 1976, i, 1203. 19. Kanis, J. A., Russell, R. G. G. Br. med. J. 1977, i, 78. 20. Haddad, J. G., Rojanasathit, S. J. clin. Endocr. Metab. 1976, 42, 284. 21. Bordier, P. J., Marie, P., Arnaud, C. D., Gueris, J., Ferriere, C., Norman, A. W. in Vitamin D and Problems Related to Uræmic Bone Disease (edited by A. W. Norman, K. Shaefer, J. W. Coburn, H. F. De Luca, D. Fraser, H. G. Grigoleit, and D. V. Herrath); p. 134. Berlin, New York, 1975. 17. 18.

ASSOCIATION OF PRESENILE CATARACTS WITH HETEROZYGOSITY FOR GALACTOSÆMIC STATES AND WITH RIBOFLAVIN DEFICIENCY

JOSEF T.

PRCHAL MARCEL E. CONRAD HAROLD W. SKALKA

Heterozygosity for galactokinase deficiency detected in 5 patients, for galactose-uridyl transferase in 2, and evidence of an erythrocytic deficiency of riboflavin in 8. Even when Black subjects were excluded from analysis because of their high incidence of polymorphism for galactokinase, these findings are significantly different from those expected from population surveys and suggest that many patients with presenile cataracts have a biochemical abnormality which can be detected by examination of red blood-cells and which may be corrected by dietary restrictions or supplements. cataracts. was

Introduction CATARACTS are a major cause of blindness all over the world.’ Identification of treatable metabolic abnormalities responsible for cataract development would be an achievement because of the morbidity of cataract surgery. Cataracts often

develop in infants with a homozygous deficiency galactokinase or galactose-1-phostransferase2 and in laboratory animals fed phate uridyl a high-galactose diet.3 Riboflavin deficiency has also been associated with cataract formation in laboratory of either

animals.4,s Abnormalities of the Leloir pathway of galactose metabolism can be identified by simple enzyme measurements in red blood-cells.2 Riboflavin deficiency has been shown to be associated with decreased activity of erythrocyte glutathione-reductase and stimulation of this enzyme by flavine adenine dinucleotide (F.A.D.) can be used as a sensitive method for its detection.6 Since homozygous galactosaemia causes cataracts in infancy, the heterozygous abnormality might be associated with cataract production later in life. Riboflavin deficiency might also be associated with an increased prevalence of cataracts in man. We have examined these hypotheses by biochemical studies of erythrocytes in patients with cataracts.

Patients and Methods 22 consecutive patients below the age of 50 who were referred to the Eye Foundation Hospital at the University of Alabama in Birmingham with idiopathic cataracts were studied. None had either a history of trauma or radiation to the eye or laboratory evidence of diabetes mellitus and hypoparathyroidism. The presence of cataracts was confirmed by slitlamp examination. Red-cell galactokinase, galactose-1-phosphate uridyl transferase, and riboflavin, were measured by the method of Beutler.7

Results Division

of Hæmatology and Oncology and Department of Ophthalmology, University of Alabama in Birmingham, Alabama 35294, U.S.A.

Red cells and the lens of the eye are nonnucleated cells; moreover, they have metabolic similarities. Cataracts develop in childhood in homozygotes for galactosæmic abnormalities, which can be detected by biochemical measurements in red bloodcells. It has not been determined whether heterozygotes for these defects are at greater risk for cataract development later in life. Similarly, riboflavin deficiency for which the erythrocyte is a sensitive indicator, has been associated with cataracts in animals. Red-cell studies were carried out in 22 patients, aged under 50, with

Summary

The accompanying table shows red-cell galactokinase and galactose-uridyl-transferase activity in 19 unrelated patients with cataract; glutathione-reductase activity with and without F.A.D. stimulation in 22 patients; and activities of each enzyme in 10 healthy volunteers who took normal diets. 5 subjects had low erythrocyte-galactokinase values in the range suggesting a heterozygous deficiency for this enzyme. The activity of galactokinase in Black subjects is significantly lower than in Caucasians.s 1 of our patients with a low value is Black (patient 1). However, even if we exclude this subject, there is a 20% frequency of heterozygotes in this small sample of patients. Although Mayes and Guthrie9 had a few Black subjects in their sample they found that heter-