In reporting the results of these investigations on the effect of the ingestion of food upon the metabolism, it seems appropriate, in view of our experience with various foods and numerous subjects during the past 12 years, to offer suggestions as to the methods to be employed for an ideal study of this problem. These suggestions are based not only upon the actual work here reported, but also upon much experimenting carried on since most of this work was done.

Objects

We must first recognize the objects of such a study. These are the determinations of (1) the total effect upon the metabolism of the ingestion of food, namely, the increase above the basal metabolism; (2) the highest increase above the basal metabolism and its time relation to the taking of food, i. e., the peak effect; and (3) the subsequent character of the metabolism to note whether it remains unaltered or if there is a change in the proportions of protein, fat, and carbohydrate metabolized.

Subjects

It is obvious that the subjects selected should be primarily normal, healthy individuals, without tendency to digestive disturbances. Only through a knowledge of the metabolism of normal individuals can we gain information as to the abnormalities in the metabolism after food ingestion under pathological conditions. Individuals who are likely to continue throughout an entire series of tests are to be preferred, as they may be depended upon for subsequent duplicate and control experiments. Repeated experiments with the same individuals obviate the necessity for training new subjects, lead to an improvement in the experimental routine, and reduce the number of subjects required for obtaining results which will supply definite conclusions. The training of pathological subjects is more difficult than the training of normal subjects; moreover, the physical condition of such subjects is liable to variation. A greater number of experiments is accordingly necessary for a series of investigations with pathological conditions.

Basal Metabolism

Since the object of any study of the metabolism subsequent to food ingestion is to determine the effect upon the basal metabolism, i. e., the metabolism in the post-absorptive condition (12 hours or more after the last food ingestion), it is necessary first to obtain an accurate measure of the basal metabolism. As our own unfortunate experience only too frequently shows, it is imperative to determine the basal metabolism and the metabolism after food upon the same day, save perhaps in exceptionally prolonged experiments. When this is not done the basal metabolism determined on another day may be higher than the true basal metabolism of the food day, thus leading to the conclusion that the effect of the food is negative. Furthermore, there should be a preliminary period of observation which should be continued 30 minutes or preferably longer, so that one may state with certainty that the basal level has been reached before the actual measurements are begun. During this preliminary period the subject should be at rest and in the same body position as during the experimental period.

Control Of External Muscular Activity

It has been repeatedly stated in publications from this Laboratory1 that, in any study of metabolism in which comparable results are to be obtained, it is necessary to have as nearly as possible complete muscular repose and that there must be a graphic record which will indicate that such repose has been maintained throughout the periods of the experiments which are compared with each other. Such a graphic record may be obtained either by means of pneumographs around the thighs and thorax or by means of a pneumograph or pneumatic bulb connected to the bed support. If these devices are connected with a sensitive recording tambour, the slightest muscular movement of the subject results in a change in the volume of the air in the pneumograph, which is immediately recorded by the pointer of the tambour upon the smoked surface of the rotating drum of a kymograph.

In addition to the record of the amount of complete external muscular repose it is equally important to note any drowsiness or sleep which may occur during the measurement of the metabolism. Recent experience with human subjects in a series of experiments upon the metabolic effect of the ingestion of alcohol has shown that the degree of wakefulness can be satisfactorily recorded by having the subject press a push button periodically in response to a stimulus. The stimulus is supplied by a signal magnet which is operated once every half minute. The magnet is so placed that the subject can hear it readily and is in series with a battery, clock, and second signal magnet which records upon a moving kymograph drum. The push button operated by the subject is connected with a battery and an independent recording signal magnet, thus giving a record of the response to the signal. A continuous record of responses gives positive evidence of wakefulness on the part of the subject, while a continuous lack of responses is indicative of drowsiness or actual sleep.

The effect of external muscular activity is to change the total metabolism, while the effect of drowsiness or sleep is to change the apparent character of the respiratory exchange; accordingly, a graphic record of both is essential for a reliable interpretation of the results obtained.

Length Of Periods

The length of the periods of observation is naturally dependent upon the total effect to be measured and upon the apparatus used. When the effect is exceedingly small, and particularly when the peak effect and its time relation are desired, it is essential to make the periods as short as possible, preferably 10 to 15 minutes. If an apparatus with a closed chamber is used, periods as short as these are not possible; with such an apparatus, periods of at least 30 minutes should be employed and 45-minute periods are more reliable.

Apparatus

From the experience of the last 12 years in studies of this character, we have come to the conclusion that some form of respiration apparatus is desirable with which it is possible to measure the gaseous exchange continuously in short periods. At present the best combination for this purpose with a trained subject is found to consist of a face mask, valves for separating inspired and expired air, two spirometers (preferably of the Tissot type), and a Haldane portable gas-analysis apparatus for analyzing the expired air. The face mask is one used in the Siebe-Gorman mine-rescue apparatus.2 To secure reliable results, the tightness of the mask against the face must be assured. The valves for separating inspired and expired air most used in this laboratory are the Thiry-Tissot model,3 but any valve which is lightly movable and gives perfect closure without backlash is suitable. The spirometers are of the Tissot model. A complete description of this spirometer is given in a former publication from this Laboratory.1 Any spirometer which is lightly movable and fairly well counterpoised is practicable for this purpose. The 100-liter spirometer is the most adaptable for general use. The portable gas-analysis apparatus is the one devised by Haldane2 for the analysis of atmospheric, mine, and expired air. Its accuracy should be controlled by analyses of atmospheric air. The apparatus, when properly set up, should give 0.03 per cent CO2 and 20.93 to 20.95 per cent O2 for atmospheric air. We insist on this as a proof of the accuracy of the analysis of the expired air. Such analyses should be reported in connection with the results of metabolism measurements.

1Benedict, Deutsch. Arch. f. klin. Med., 1912, 107, p. 156; Benedict and Talbot, Carnegie Inst. Wash. Pub. No. 201,1914, pp. 31 and 59; Benedict, Carnegie Inst. Wash. Pub. No. 203,1915, p. 311.

2This may be obtained from H. N. Elmer, 1140 Monadnock Building, Chicago, 111.

3Both the Thiry-Tiasot valves and Tissot spirometer were obtained from Pirard and Coeurde-vache, 7 rue Blainville, Paris, France.

With the combination of apparatus outlined in the foregoing paragraphs, a trained subject awake, and a complete absence of external muscular activity, it is possible to measure the peak effect of either the metabolism or of the respiratory quotient, or to measure the effects of the ingestion of exceedingly small amounts of material. During the past two years the gaseous exchange and respiratory quotients of trained subjects have been measured for 6 to 7 hours in consecutive experimental periods as short as 10 minutes, with no great degree of discomfort to the subject and with a high degree of accuracy.

When the increments in metabolism are likely to be large and to extend over a considerable period of time, and when it is possible to repeat the experiment several times, the clinical respiration apparatus (a chamber apparatus)3 is probably the most feasible. In this apparatus it is not necessary for the subject to remain absolutely immovable and the possibility of movement makes it pleasanter for the subject in long experiments.

Summary

From the foregoing it can be seen that the ideal method for determining the effect of the ingestion of food upon the metabolism is the use of trained subjects; a measurement of the basal metabolism on the same day as that following the ingestion of food; an absolute absence of external muscular activity; a subject awake; graphic records of both the absence of activity and any evidence of drowsiness or sleep; as short periods as possible; spirometers, respiratory valves, face mask, and portable Haldane gas-analysis apparatus, with determinations controlled by analyses of atmospheric air.

1Carpenter, Carnegie Inst. Wash. Pub. No. 216, 1915, p. 61.

2Haldane, Methods of air analysis, 1912, p. 47.

3Benedict and Tompkins, Boston Med. and Surg. Journ., 1916, 174, pp. 857, 898, and 939.