205 by the pancreas. Measurement of this activity the homogenate (Goldberg, Campbell and Roy, may then provide an index of pancreatic function to be published). Results on two subjects before more complete and technically less demanding and after ileostomy indicate that the fecal T.P.A. than estimation of the activity of individual is considerably less than that of the ileum, so that pancreatic enzymes. Such estimations should be inactivation in the colon seems likely. based upon the total output of the subject on a controlled protein intake, and should be made Conclusions on a homogenate of the material. Centrifugation The most striking finding is the remarkable of a homogenate at 3,000 g sediments about 30 % of the T.P.A. and this is due to adsorption of capacity of the human small intestine for protein trypsin, chymotrypsin and possibly other digestion. When allowance is made for intestinal pancreatic enzymes by cellular debris present in transit time and for the temperature of the intestinal contents (37°), it is apparent that several kg. of protein can be digested per day, PROTEOLYTIC ACTIVITY assuming that the acid-soluble fragments as dv.ID TRYPSIN+ CHYMOTRYPSIN measured in the T.P.A. assay are suitable for 750 absorption. Thus, in human disease, loss of protein is more likely to result from intestinal hurry, from excessive endogenous secretion, or from failure of amino-acid absorption, than 500 through failure of protein digestion. r=0·708

t=1S·92 250

0'--------,------.,.-----..----, 500

750

1000

TOTAL PROTEOLYTIC ACTIVITY

Fig.6.-Regression Line for T.P.A. on Proteolytic Activity due to Trypsin and Chymotrypsin Content of Sample (254 specimens).

References Folin, O. and Ciocalteau, V. (1927), J. bioi. Chem., 73, 627. Fruton, J. S. (1961), p. 233 "The Enzymes," Vol. 4. Ed. Boyer, P. D., Lardy, H. and Myrback, K. Academic Press, New York. Gowenlock, A. H. (1953), Biochem. J., 53,274. Haverback, B. J., Dyce, B. J., Gutentag, P. J. and Montgomery, D. W. (1963), Gastroenterology, 44,588. Hunt, J. N. (1948), Biochem. J., 42, 104. Kunitz, M. (1947), J. gen. Physiol., 30, 291.

Serum Lipid Levels following a Fatty Meal as a Test of Steatorrhoea W. A. F. PENFOLD DEPARTMENT OF CLINICAL BIOCHEMISTRY, UNITED OXFORD HOSPITALS, OXFORD

The presenting symptoms of malabsorption are so varied that the search for the primary fault often involves prolonged and detailed investigations. Many tests have been found to aid the diagnosis, including blood urea, serum electrolytes, proteins and iron, folate, B 12 hemoglobin estimations, blood film examinations, gastric function tests and xylose tolerance tests. Although these, are of great importance, the established method of assessing steatorrhoea is by fecal fat determination, either as a fat balance experiment or on several days collection with dietary fat control. The disadvantages of the test are well known. The collection of the fecal specimens is un-

pleasant not only to the patient but also to the nursing staff and the actual determination is distasteful, if not, at times, positively revolting. The major disadvantage, however, is the difficulty of ensuring that the collections are complete and represent the period under study. There is reason to believe that of 1,085 fecal fat determinations, which we have carried out during the past three years, the collections may have been incomplete or unrepresentative in as many as 30 % of the cases. Many papers and publications dealing with fat excretion indicate the wide day-to-day variation in the amount of fat excreted by normal as well as sick people. There is need for pro-

206 longed collection of feeces with dietary control, if reliable and true fecal fat excretion rates are to be obtained. Recently, it has become the custom in many hospitals to do fecal fat determinations on shorter periods-often four or five days-without dietary fat control. In our hospital, a collection period of four days is considered the minimum for a reliable fat absorption assessment. Although this permits great saving in patients' "bed stay" time, much less work for the dieticians and for the laboratory, if means that the normal range is less well defined. An efficient and reliable means of assessing fat absorption without handling of faces would be welcome. The substitution of a test based upon the estimation of serum lipids in place of the fecal fats would eliminate the unpleasantness of the collection and handling of faces as well as the cleansing of the containers and apparatus used in the determinations. It would permit a much quicker result to be obtained. Many workers have attempted this by studying the blood lipids, but the normal ranges of lipids in the randomly collected blood sample or the fasting sample are so wide as to be useless for this purpose.

Methods The serum fatty acids were determined by the procedure of Stern and Shapiro (1953) and the facal fats by the method of Van de Kamer et al (1949). Result-Normal subjects The serum lipids, before and after the meal have been studied in 20 healthy controls and the range of normal results are outlined in Fig. I. The fasting levels ranged from 255-440 mg.j100 ml, and in all cases the serum lipids showed a marked rise after the meal, the minimum rise being 88 mg.jlOO ml. either in the two or four hour specimen. The results of duplicate analyses on 12 subjects, normal controls and patients, are given in Table 1. The duplicates agreed well, giving a standard deviation of ±4.5 at fasting level, and ± 7.0 mg.jlOO ml. at the peak. SERUM

,"I/IOOm l

In this paper we report our attempt to assess fat absorption by the study of the rise in the blood lipids following a standardised fatty meal. The changes in the level of fatty acids in serum taken from patients after an oral dose of fat in the form of a meal of known fat content, have been measured, and results correlated with the daily fecal fat excretion. Subjects Studies were performed in 55 subjects; 20 laboratory staff as controls and 35 patients being investigated for mal-absorption syndrome. Each subject continued on a normal diet until the evening prior to the morning of the test. After an overnight fast, a specimen of venous blood was collected and the serum separated. A high fat meal containing 100 g. fat supplied in 4 ozs. cream, 2 ozs. butter, 2 ozs. bread (as toast), 3 ozs. banana, was eaten; half a pint of tea or coffee was also given, saccharin being allowed as a sweetening medium. Further blood specimens were collected at 2 and 4 hour intervals, and the total serum fatty acids estimated. The meal was well tolerated by aU the subjects, and although some patients did not eat all the butter, this did not appear to influence the results significantly.

II PIDS (abnormal)

~:411lJ11II111II1Il1111II1II11IJ11I"1II111111

~~~

~:~

--

"Il!!l11I11I1111111I11I11IIIIIII""1111I11I11

£,

F

2 hr

4hr

Fig. I.-Serum Lipids (mg.jlOO ml.), before and after test meal, of patients with Steatorrhoea. (Ranges of serum lipids of normal controls outlined by shaded lines).

207

TABLE I Highest Point After Meal

Fasting

1 2 3 4

5 6 7 7 8 9 9 10 11

12

210 236 260 310 350 442 364 364 440 272 278 370 255 330

216 236 262 312 346 435 360 436 280 370 260 325

240 252 270 350 460 546 475 478 590 360 372 570 380 500

252 252 273 347 460 540 415 580 372 565 388 495

Mean of results. (Normal cases - 5 to 12, Abnormal cases I to 4) Normal 346.5 351.5 473.1 483.3 256.5 Abnormal 254 278 281

Increase

30 16 10 40 110 104 111 114 150

36 16 11 35 114 105 115

88

144 92

94 200 125 170

195 128 170

126.6 24.0

135.3 24.5

Serum Lipids (mg./100 ml.) Repeated Tests (Patients No.7 and 9-triplicate tests) Triplicate test meals were given to patients seven and nine (Table I) showing that over a limited period of time the response to a fatty meal appears constant, and that the condition of the test gives good reproducibility. Recovery experiments of added fatty acids resulted in recoveries of l00±3.0%. As the result of our study of control subjects, we have accepted as the normal ranges-a rise of more than 90 mg./lOD mi. in the serum lipids after the meal and a fecal fat excretion of less than 5.6 g./24 hr. Patients Thirty-five patients were investigated for possible malabsorption. Of these, II were proved not to have it, their fecal fat excretion was within the normal range and varied from 1.8 to 4.7 g.1 24 hr. The serum lipids, before and after the meal were also normal. The fasting levels ranged from 315 to 515 mg./lOO mi. serum, a slightly wider range than the healthy controls and in every

case the rise following the meal was greater than 90 mg./lOOmi. (-the range being 95 to 165 mg.), In the remaining 24 patients, the fecal fat estimations indicated impaired absorption of fat. The serum lipids, before and after the meal, are shown in Fig. I and the normal range is shown by the shaded lines. Although in most cases the fasting levels were within the normal range, in many cases no significant rise after the meal was detected, and in the others only a minimal rise of 6 to 45 mg./lOOml. In these cases there was a clear cut distinction between the normal and abnormal patient. Three cases warrant special mention, the first, a lady suffering from ceeliac disease-and who is gluten-sensitive. The results of three tests (a) done before she was on gluten-free diet, (b) when she was taking this diet, and (c) when she had ceased to take the diet, are shown in Fig. 2. The fall and rise of the fecal fat excretion contrasts strongly with the rise and fall of the serum lipids following a test fatty meal-the normal fecal fat shown labelled (b) is associated with a normal rise in the serum lipids.

208 GLUTEN

The second case was a man complaining of indigestion, in whom fsecal fat determination showed fat excretion of 14.8, 10.6 and 16.0 g./ 24 hr. The rise in lipids following a meal were 27, 32 and 23 mg./IOO ml. (serum). Subsequent investigation demonstrated a carcinoma of the colon. Two cases of Crohn's disease have also shown an increased fat excretion and an abnormally low serum lipid rise following the meal. However, one case did not fit into our general findings. This was a man of 53 years with a long standing chronic pancreatitis with diabetes. Repeated tests showed an abnormal fat excretion, amounting to 12, 11 and 10 g./24 hr. on three separate occasions. The rise in the serum lipids, after the meal, however, were within the normal range, namely 175, 138 and 115 mg./l00 ml, Another case of chronic pancreatitis has not behaved like this and has shown the usual increased fat excretion associated with the decreased serum lipid rise. We can offer no explanation of the behaviour of this case. The results of this investigation are summarised in Table II and Fig. 3. The normal controls and the patients without malabsorption had a fecal fat excretion ofless than 5.6 g./24 hr. and a rise in the serum lipids, following the meal, of more than 90 mg./l00 ml, The patients with malabsorption all had a fat excretion of more than 5.8 g./24 hr. and a rise in the serum lipids ofless than 50 mg./IOO ml. (Table II).

SENSITIVE

PATI E NT

mgl 100 ml b

hecal fu

I/Hhr s 30 c 20

200

a

I F

10

• b

Z hr

4hr

Fig. 2.-Serum lipids (mg.{lOO ml.) and faecal fat excretion of "gluten sensitive" patient. (Patient on-{a) normal diet, (b) giuten-free diet, (e) normal diet).

Summary

A comparative study of the rise in serum lipids following a standardised fatty meal and the daily

TABLE

II

Frecal Fat (g./24 hrs.)

Serum Lipids (mg./IOO ml.) I

Fasting Range

Rise after Meal Mean Range

Range

Mean

Control Group

255-440

88-250

151

1.6-5.5

3.2

Normal Patients

315-515

95-165

139

1.8--4.7

3.4

6--45

24

5.9---62

17.0

Abnormal Patients 180--520

Summary of Results

209

-

I

t ..:._-- -_..:_._- -.--250 200

150 100

50

•

fecal fat excretion has shown that, in 98 % of a group of normal subjects and patients with malabsorption, a normal fat excretion is associated with a rise in the serum lipids of more than 90 mg./lDO ml, after the meal and an abnormal fecel fat excretion with a rise of less than this and that the assessment of malabsorption by means of the serum lipid rise test can be obtained in less than one day compared with four to six days when the assessment is based upon fsecal fat studies.

Faecal fats (g./24 hr.) 0

Acknowledgement I thank Mr. J. R. P. O'Brien, in whose department this work has been carried out and the Physicians of The United Oxford Hospitals for permission to study their patients. References Kamer, J. H. van de, Huinink, H., and Weyers, H. A. (1949), J. biol. Chern.• 177, 347. Stem, I. and Shapiro, B. (1953), J. clin. Path.• 6, 158.

SOUTHERN ENGLAND AND SOUTH WALES REGION

Meeting held in London, February, 1967 Intestinal Absorption of Vitamin B1 2 and Folic Acid D. M. MAITHEWS DEPARTMENT OF CHEMICAL PATHOLOGY, VINCENT SQUARE LABORATORIES OF WESTMINSTER HOSPITAL, LONDON

The absorption of vitamin Bl2 cannot be dealt with adequately in any short paper. The aim of this paper is to give a concise account of modem views on the subject, necessarily oversimplified, and to elaborate a few points which the author regards as particularly interesting. When no reference is quoted, the reader is referred to the following: Herbert, 1959; Mollin, 1959; Grasbeck, 1960; Wilson, 1962, 1964;

Vitamin Bl2 und Intrinsic Factor, 1962; Wiseman, 1964; Lester Smith, 1965; Matthews, 1967. Vitamin Bl2 in food is virtually confined to animal products, and is mainly protein-bound. It occurs in several forms-c-eo-enzyme Bl2 • methylcobalamin, hydroxocobalamin and cyanocobalamin, co-enzyme Bl2 probably predominating. Most of the work on absorption has been