chapter I

shall only give single methods or point out short cuts in well-recognized ones or make suggestions as to new methods of blood study which may eventually aid us in diagnosis.

Those who work in temperate climates cannot realize the difficulties which beset the tropical laboratory worker from the lack of proper assistance, damaging effects of heat and moisture on stains and media and, of greater importance, the impairment of that driving energy so necessary for the carrying out of complicated methods. A short and simple method has a far greater value in the tropics than at home.

BLOOD PREPARATIONS

To obtain blood, except for blood cultures, use either a platino-iridium hypodermic needle which can be sterilized in the flame, a small tenotome, or a surgical needle with cutting edge.

Needles should be sterilized by boiling since flaming dulls the edge. A steel pen with one nib broken off or the glass needle of Wright may also be used. To make a glass needle, pull straight apart a piece of capillary tubing in a very small flame. Tap the fine point to break off the very delicate extremity. Scarcely any pain attends the use of such a needle. In puncturing either the tip of the finger or lobe of the ear a quick piano-touch-like stroke should be used. The ear is preferable, as it is less sensitive and there is less danger of infection. Before puncturing, the skin should be cleaned with 70% alcohol and allowed to dry. It is advisable to sterilize the needle before using it.

Note that in order to secure in the specimen a cell count that corresponds to that obtaining in the circulation as a whole it is necessary to massage the ear vigorously prior to making the puncture. Subsequently there should be no manipulation of the part, the blood examined being that which exudes freely. This procedure renders more likely the finding of blood parasites.

The first drop of blood which exudes should be taken up on the paper of the Tallquist haemoglobinometer, using subsequent ones for the blood pipettes and smears. If it is necessary to make a complete examination, it is rather difficult to draw up the blood in the pipettes, dilute it, and then get material for fresh blood preparations and films without undue squeezing, which is to be avoided. Of course, fresh punctures can be made. Ordinarily, complete blood examinations are not called for. It is only a white count or a differential count or an examination for malaria that is required.

As a practical point it is very rare that a red count is indicated. There is one point not sufficiently recognized by physicians and that is that a routine blood examination is not apt to be as carefully conducted as one calling for a specific feature. Without disparaging the necessity of routine examination of urine as well as blood it is a fact that the internist who knows what he wants gets better results from the laboratory man.

THE MICROSCOPICAL EXAMINATION OF FRESH PREPARATIONS OR STAINED BLOOD SMEARS

As regards haemocytometry it may be stated that in the tropics the counting of red cells is required more frequently in comparison to white ones than is the case in temperate climates where probably 100 white counts are necessitated as against 1 red count. This is on account of the frequency of secondary anaemias in the tropics.

The idea that time may be saved by making a white and red count from the same preparation is not borne out practically so that it is better to make white and red counts separately.

As a diluting fluid for red counts a normal salt solution, preferably about 0.9%, answers perfectly and if desired may be tinged with neutral red, methyl green or gentian violet to bring out white cells. When available, however, I prefer a 2½% aqueous solution of potassium bichromate for red cell counts.

=Rulings.=—The most desirable rulings are those of Türck, Zappert and Neubauer.

[Illustration: FIG. 147.—Neubauer’s ruling.]

In these the entire ruled surface consists of nine large squares, each 1 mm. square. These are subdivided, and in the central large square are to be found the small squares used for averaging the red cells. These small squares are 1/20 mm. square and are arranged in nine groups of 16 small squares by bordering triple-ruled lines. As the unit in blood counting is the cubic millimeter, if one counted all the white cells lying within one of the large squares (1 mm. square), he would have only counted the cells in a layer one-tenth of the required depth, so that it would be necessary to multiply the number obtained by 10. This product, multiplied by the dilution of the blood, would give the number of white cells in a cubic millimeter of undiluted blood. The Neubauer ruling is the most satisfactory.

[Illustration: FIG. 148.—Thoma-Levy counting chamber, Bürker double type with two Neubauer rulings.]

=Bürker Haemacytometer.=—Some workers prefer the _Bürker haemacytometer_. In this there are two ruled wedge-shaped pieces of glass, separated at their bases, which take the place of the ruled disc of the Thoma apparatus. Two oblong pieces of glass are on either side of the ruled wedges and are 0.1 mm. higher, thus taking the place of the shelf. Clamps fix a cover-glass on these shelves giving a space 1/10 mm. over the ruled surfaces. The blood is run in by capillarity from the mixing pipette. I gave up this type of counter because the clamps made manipulation awkward.

=Thoma-Levy Chamber.=—In the Thoma-Levy modification of the Bürker apparatus the central portion of the slide is cut away and in this depression is cemented a rectangular strip of glass, divided by a central channel. Each half of this strip of glass has a Neubauer ruling on it so that one can make a white count on one side and a red one on the other, simply touching the tip of the red pipette to the space separating the under surface of the cover-glass from the ruled rectangular slips on one side and then with the white pipette repeating the same on the other side. An advantage of the Thoma-Levy is that the original thickness of the slide makes the shelf on which the cover-glass rests instead of the support being strips on either side of the ruled surfaces and cemented to the slide. The Neubauer ruling is undoubtedly the most satisfactory of the haemacytometer rulings, its rulings being simpler than those of the Türck system. The unit square in all these haemacytometers is the small square for counting red blood cells, 1/20 mm. square.

[Illustration: FIG. 149.—The Türck ruling. Thoma-Zeiss Haemacytometer.]

=To Make a Red Count.=—Having a fairly large drop of blood, apply the tip of the 101 pipette to it and, holding the pipette horizontal, carefully and slowly draw up with suction on the rubber tube a column of blood to exactly 0.5. The variation of 1/25 of an inch from the mark would make a difference of almost 3%. If the column goes above 0.5, it can be gently tapped down on a piece of filter-paper until the 0.5 line is cut. Now insert the tip of the pipette into some diluting fluid, and revolving the pipette on its long axis while filling it by suction, you continue until the mark 101 is reached.

A variation of 1/25 of an inch at this mark would only give an error of about 1/30 of 1%. This gives a 1-200 dilution. After mixing thoroughly by shaking for one or two minutes, the fluid in the pipette below the bulb is expelled (this of course is only diluting fluid). A drop of the diluted blood of a size just sufficient to cover the disc when the cover-glass is adjusted, is then deposited on the disc and the cover-glass applied by a sort of sliding movement, best obtained by using forceps in one hand assisted by the thumb and index-finger of the other.

In red counts we use exclusively the small 1/20 mm. squares which are in groups of 16 bounded by triple-ruled lines.

The depth of fluid over the ruled surface is 1/10 mm., hence each of these small squares is 1/10 × 1/20 × 1/20 = 1/4000 of a c.mm., so that it takes 4000 such spaces to equal the unit for blood counting (1 c.mm.). My practice in making red counts is to count the red cells in five of the groups of 16 small squares. This in normal blood is about 100 for the 16 squares. After counting 5 groups of 16 we have counted the red cells of 80 small squares which is 1/50 of 4000 (the number in the 1 c.mm. unit). For this reason 50 × 200 (the blood dilution) = 10,000, so that it is only necessary to multiply the number of red cells found in 5 groups of 16 small squares by 10,000 in order to obtain the number of red cells per c.mm. For more accurate determination the process can be repeated with a second or third drop of the diluted blood, which would give an average from 160 or from 240 small squares.

=To Count White Cells.=—Draw up the blood in the white pipette to the 0.5 line. Then, still holding the pipette as near the horizontal as possible, because the column of blood tends to fall down in the larger bore, draw up by suction a diluting fluid which will disintegrate the red cells without injuring the whites. The best fluid is 0.5% of glacial acetic acid in water. This makes the white cells stand out as highly refractile bodies. Some prefer to tinge the fluid with neutral red or gentian violet. The 0.5 mark is preferred because it takes a very large drop of blood to fill the tube up to the 1 mark and if there is much of a leucocytosis a 1 to 10 dilution is not sufficient.

The blood having been drawn up to 0.5, we have a dilution of 1 to 20.

Making a preparation, exactly as was done in the case of the red count, we count all of the white cells in one of the large squares (1 sq. mm.). The cross ruling greatly facilitates this. Note the number. Then count a second and a third square. Strike an average of the large squares counted and multiply this by 10, as the depth of the fluid gives a content equal to only 1/10 of a c.mm. Then multiply by the dilution.

EXAMPLE.—First large square 50; second large square 70; third large square 60. Average 60. Then 60 × 10 × 20 = 12,000, the number of leucocytes in 1 c.mm. of blood. In order to save time the count is preferably made with a low power (⅔-inch objective) as the leucocytes stand out like pearls. It is more accurate, however, to use a higher power, so that pieces of foreign material may be recognized and not enumerated as white cells.

If one will accustom himself to comparing the distribution of the leucocytes in a well-made stained dried-blood film, prepared according to Ehrlich’s cover-glass method, with that in a haemacytometer preparation, he can readily acquire an experience which will enable him to determine with considerable accuracy the degree of leucocytosis by the examination of a stained, cover-glass preparation alone.

After making a blood count, the haemacytometer slide should be cleaned with soap and water and then rubbed dry, preferably with an old piece of linen. As the accuracy of the counting chamber depends upon the integrity of the cement, any reagent such as alcohol, xylol, etc., and in particular, heat, will ruin the instrument. The pipettes should be cleaned by inserting the ends into the tube from a vacuum pump, as a Chapman pump. First draw water or 1% sod. carbonate solution through the pipette, then alcohol, then ether, and finally allow air to pass through to dry the interior. If the interior is stained, used 1% HCL in alcohol. If a vacuum pump is not at hand, a bicycle pump or suction by mouth will answer.

PREPARATIONS FOR THE STUDY OF FRESH BLOOD

Many authorities prefer a fresh blood specimen to a stained dried smear in the study of parasites of the blood. In malaria in

## particular there is so much information as to species to be obtained

from a fresh specimen that the employment of this method should never be neglected. While waiting for the film to stain one has five or six minutes which could not be better spent than in examining the fresh specimen which only requires a moment to make.

=Manson’s Method.=—Have a perfectly clean cover-glass and slide. Touch the apex of the exuding drop of blood with the cover-glass and drop it on the center of the slide. The blood flows out in a film which exhibits an “empty zone” in the center. Surrounding this we have the “zone of scattered corpuscles,” next the “single layer zone” and the “zone of rouleaux” at the periphery. It is well to ring the preparation with vaseline. When desiring to demonstrate the flagellated bodies in malaria, it is well to breathe on the cover-glass just prior to touching the drop of blood.

=The Method of Ross= is very easy of application and gives most satisfactory preparations. Take a perfectly clean slide, and make a vaseline ring or square of the size of the cover-glass. Then, having taken up the blood on the cover-glass, drop it so that its margin rests on the vaseline ring. Gently pressing down the cover-glass on the vaseline makes beautiful preparations which keep for a very long time. If it is desired to study the action of stains on living cells, this method is also applicable. A very practical way to do this is to tinge 0.85% salt solution containing 1% sodium citrate (the same as is used in opsonic work) with methylene azur, gentian violet, or methyl green. With a capillary bulb pipette, take up one part of blood, then one part of tinted salt solution. Mix them quickly on a slide and then deposit a small drop of the mixture in the center of the vaseline ring and immediately apply a cover-glass and press down the margins as before. This method will be found of great practical value.

PREPARATION AND STAINING OF DRIED FILMS

When preparations are desired for a differential count, Ehrlich’s method of making films is to be preferred, as the different types of leukocytes are more evenly distributed. In making smears by spreading, there is a tendency for the polymorphonuclears to be concentrated at the margin while lymphocytes remain in the central part of the film.

=Cover-slip Films.=—In _Ehrlich’s method_ we have perfectly clean dry cover-slips. Take up a small drop of blood without touching the surface of the ear or finger. Drop this cover-glass immediately on a second one and as soon as the blood runs out in a film, draw the two cover-slips apart in a plane parallel to the cover-glasses. Ehrlich uses forceps to hold the cover-glasses to avoid moisture from the fingers, but I find I can work more quickly and satisfactorily with the fingers alone. The method shown in Fig. 150 is a very convenient one. In making malarial smears it is better to wash the finger or ear with soap and water to get rid of all grease and dirt. Then dry thoroughly before puncturing. Alcohol is not so efficient.

[Illustration: FIG. 150.—1, 2, 3, 4, Making blood smears on slide. 5. Smear ready for staining—grease marks prevent Wright stain from running over slide. 6. U-shaped glass tubing to hold slide in staining. 7. Right hand holding two cover-glasses. One cover-glass is being touched to drop of blood from ear. 8. Cover-glasses transferred to left hand in preparing to place one cover-glass on another and spread film. 9. Separating cover-glasses by sliding one from the other.]

Slides and spreaders should be absolutely clean and grease-free. Scrubbing with soap and water, thorough rinsing and drying, then subjecting the slide to the flame to make it grease-free is satisfactory.

For removing dirt and grease from skin, a mixture of acetone, 40; alcohol, 60; is the best and quickest means. A bottle is kept on hand, with the puncturing needle embedded in the stopper.

For cleaning a slide, nothing equals Bon Ami. Rub up some with the wet finger, rub the slide with the lather until there is a friction squeak; let dry; polish with a clean, dry cloth. This is far better than soap and water, alcohol, ether and flaming combined. Note how a drop of water spreads on a glass so treated.

=Smears on Slides.=—Of the various methods of spreading films on slides, that described by Daniels is quite satisfactory. In this the drop of blood is drawn along and not pushed along. The films are even, can be made of any desired thickness by changing the angle of the drawing slide, and there is little liability of crushing pathological cells. Take a small drop of blood on the end of a clean slide. Touch a second slide, about ½ inch from end, with the drop and as soon as the blood runs out along the line of the slide end, slide it at an angle of 45° to the other end of the horizontal slide. The blood is pulled or drawn behind the advancing edge of the advancing slide. An angle less than 45° makes a thinner film; one greater, a thicker film.

Instead of a slide a square cover-glass may be used and if the edge be smooth it makes a more satisfactory spreader than the slide.

Instead of the Daniels method I prefer to take up the drop of blood on the slide on which the smear is to be made, about ½ inch from the end. Then apply the spreader slide and so soon as the drop runs along the end of the spreader slide proceed as above described. This method is shown in Fig. 150.

=Spreaders.=—Of the various methods of making smears by means of cigarette paper, rubber tissue, needles, etc., the best seems to be to take a piece of capillary glass tubing and use this instead of a needle in making the film. There is one advantage about the strip of cigarette paper touched to the drop of blood and drawn out along the slide or cover-glass, and that is that it is almost impossible not to make a working preparation by this method.

THICK-FILM METHODS

Such methods are of the greatest practical value in searching for malarial parasites when they are in very small numbers in the peripheral circulation, in finding trypanosomes, relapsing fever spirochaetes and filarial embryos. Ruge’s method so brings out the polymorphonuclears that such a technic can be used for opsonic index. Many workers prefer the _Ross thick-film method_ in examining for malaria. In this about one-half of a drop of blood is smeared out over a surface about equal to that of a square cover-glass and allowed to dry. It is then flooded with 1/10 of 1% aqueous solution of eosin for about fifteen minutes. The preparation is then gently washed with water and then treated with a polychrome methylene-blue solution. After a few seconds this is carefully washed off and the preparation dried and examined.

James smears out an ordinary drop of blood so that it makes a circular smear about ¾ inch in diameter. This may be easily accomplished with a spatulate toothpick. When dry, treat the blood smear with alcohol containing HCl (Alcohol 50 cc., HCl 10 drops) until the haemoglobin is dissolved out. Then wash thoroughly in water for five or ten minutes. Allow to dry and then stain as ordinarily with the Wright or Giemsa stain.

=Ruge’s Method.=—The best thick-film method is that of Ruge. After the blood has dried well gently move the slide about in a glass containing a 2% solution of formalin to which has been added 1% of glacial acetic acid. After laking is completed, as shown by disappearance of brown color, treat the slide in the same way in a glass of tap water to remove all traces of acid. Next wash gently in distilled water and stain with dilute Giemsa (1 drop to 1 cc. of water) for twenty to thirty minutes. Wash in water and allow to dry without heat or blotting paper. Some workers prefer to stain the dried thick smear for one hour in a jar containing dilute Giemsa stain (1 to 40) without previous fixation or dehaemoglobinization. At present, I make my thick films by taking up a large loopful from the exuding drop of the puncture wound.

This is deposited at one end of the slide and from it three or four more daubs are made in succession toward the other end of the slide. These daubs are quickly smeared out before coagulation takes place in the first daub.

With all thick-film methods it is extremely important to have thorough drying of the smear before dehaemoglobinizing or staining. This ordinarily requires one or two hours in the air or twenty to thirty minutes in the incubator. It is particularly important in working with such smears, although holding for ordinary smears, to protect them from flies, ants, etc., as such insects will eat up the smear in a few minutes if left exposed.

=Fixation of Film.=—In Wright’s, Leishman’s, and other similar stains the methyl-alcohol solvent causes the fixation. In staining with Giemsa’s stain, or haematoxylin and eosin, separate fixation is necessary. For Giemsa either absolute alcohol (ten to fifteen minutes) or methyl alcohol (two to five minutes) answers well.

For haematoxylin and eosin, heat gives the best results. The best method is to place the films in an oven provided with a thermometer. Raise the temperature of the oven to 135°C. and then remove the burner. After the oven has cooled, take out the fixed slides or slips.

One of the handiest methods is to drop a few drops of 95% alcohol on the slide or cover-glass. Allow this to flow over the entire surface; then get rid of the excess of alcohol by touching the edge to a piece of filter-paper for a second or two. Then light the remaining alcohol film from the flame and allow the burning alcohol to burn itself out.

=Staining Blood-films.=—As separate staining with eosin and methylene blue rarely gives good preparations and as the modifications of the Romanowsky stain recommended are easy to make and employ, and give much greater information, the separate method of staining is not recommended.

_Wright’s Method._—The stain is made by adding 1 gram of methylene blue (Grubler) to 100 cc. of a ½% solution of sodium bicarbonate in water. This mixture is heated for one hour in an Arnold sterilizer. The flask, containing the alkaline methylene-blue solution should be of such size and shape that the depth of the fluid does not exceed 2½ inches. When cool, filter the methylene blue solution, and add 500 cc. of a 1 to 1000 eosin solution (yellow eosin, water soluble). Add the eosin solution slowly, stirring constantly until the blue color is lost and the mixture becomes purple with a yellow metallic lustre on the surface, and there is formed a finely granular black precipitate. Collect this precipitate on a filter-paper and when thoroughly dry (dry in the incubator at 38°C.) dissolve 0.3 gram in 100 cc. of pure methyl alcohol (acetone-free). Wright lately has recommended using 0.1 in 60 cc. methyl alcohol. This constitutes the stock solution. For use filter off 20 cc. and add to the filtrate 5 cc. of methyl alcohol.

A _modification by Balch_ is very satisfactory. In this method instead of polychroming the methylene blue with sodium bicarbonate and heat, the method of Borrel is used. Dissolve 1 gram of methylene blue in 100 cc. of distilled water. Next dissolve 0.5 gram of silver nitrate in 50 cc. of distilled water. To the silver solution add a 2 to 5% caustic soda solution until the silver oxide is completely precipitated. Wash the precipitated silver oxide several times with distilled water. This is best accomplished by pouring the wash-water on the heavy black precipitate in the flask, agitating, then decanting and again pouring on water. After removing all excess of alkali by repeated washings, add the methylene-blue solution to the precipitated silver oxide in the flask. Allow to stand about ten days, occasionally shaking until a purplish color develops. The process may be hastened in an incubator. When polychroming is complete, filter off and add to the filtrate the 1 to 1000 eosin solution and proceed exactly as with Wright’s stain.

In _Leishman’s method_ the polychroming is accomplished by adding 1 gram of methylene blue to 100 cc. of a ½% solution of sodium carbonate. This is kept at 65°C. for twelve hours and allowed to stand at room temperature for ten days before the eosin solution is added. The succeeding steps are as for Wright’s stain.

_In all Romanowsky methods_ distilled water should be used. If not obtainable, the best substitute is rain-water collected in the open and not from a roof.

_Method of staining:_

1. Make films and air dry.

2. Cover dry film preparation with the methyl-alcohol stain for one minute (to fix).

3. Add water to the stain on the cover-glass or slide, drop by drop, until a yellow metallic scum begins to form. It is advisable to add the drops of water rapidly in order to eliminate precipitates on the stained film. Practically, we may add 1 drop of water for every drop of stain used.

4. Wash thoroughly in water until the film has a pinkish tint.

5. Dry with filter-paper and mount.

Red cells are stained orange to pink; nuclei, shades of violet; eosinophile granules, red; neutrophile granules, yellow to lilac; blood platelets, purplish; malarial parasites, blue; chromatin, metallic-red to rose-pink.

_Giemsa’s Modification of the Romanowsky Method._—This is one of the most perfect of the modifications. The objection is that greater time in staining films is required than with the Wright or Leishman method and the stain is very expensive.

Take of Azur II eosin 0.3 gram. Azur II 0.08 gram.

Dissolve this amount of dry powder in 25 cc. of glycerine at 60°C. Then add 25 cc. of methyl alcohol at the same temperature. Allow the glycerine-methyl alcohol solution to stand overnight and then filter. This is the stock stain. To use: Dilute 1 cc. with 10 to 15 cc. of distilled water. If 1 to 1000 potassium carbonate solution is used instead of water it stains more deeply. These same dyes, mixed with methylene violet, are now obtainable commercially as a powder ready for solution in methyl alcohol.

The alkaline diluent is used to obtain the coarse stippling in malignant tertian (Maurer’s clefts). Having fixed the smear with methyl alcohol for one to five minutes, pour on the diluted stain, and after fifteen to thirty minutes wash off and continue washing with distilled water until the film has a slight pink tinge. For _Treponema pertenue_ stain from one to twelve hours.

=Haematoxylin Staining.=—While the Romanowsky methods are more satisfactory for differential counts and for the demonstration of the malarial parasites, and especially for differentiating species, yet by reason of the liability to deterioration in the tropics of methylene blue the haematoxylin methods may be preferable. Many workers in blood-work and cytodiagnosis prefer the haematoxylin.

1. Fix the film either by heat, with methyl alcohol for two minutes or with Whitney’s fixative. Heat is to be preferred.

2. Stain with Meyer’s hemalum or Delafield’s haematoxylin for from five to fifteen minutes according to the stain. Frequently three minutes will be found sufficient. To make the hemalum, dissolve 0.5 gram of haematin in 25 cc. of 95% alcohol. Next dissolve 25 grams of ammonia alum in 500 cc. of distilled water. Mix the two solutions and allow to ripen for a few days. The stain should be satisfactory in two or three days.

3. Wash for two to five minutes in tap water to develop the haematoxylin color.

4. Stain either with a 1 to 1000 aqueous solution of eosin or with a one-half of 1% eosin solution in 70% alcohol. The eosin staining only requires fifteen to thirty seconds.

5. Wash and examine.

DIFFERENTIAL COUNT

In making a differential count I would recommend the following from the directions of Schilling-Torgau. It will be remembered that considerable interest was raised a few years ago in what was termed the Arneth index. In this the more normal, more mature, better resisting polymorphonuclears were considered to have 3 or 4 lobes to the nuclear structure, even occasionally 5. The immature cells had only one or at most two lobes to the nucleus. The index was obtained by adding the percentages of cells showing 1 and 2 lobes to ½ the percentage of those with 3 lobes. As will be understood a high percentage of these immature cells was unfavorable in prognosis. These cells are graded from left to right, I, II, III, IV, V, as to separate masses in the nucleus, so that when the percentage is shoved or displaced to the left it indicates an increase in the immature cells.

Schilling-Torgau divides his polymorphonuclears into: (1) The myelocyte which is always of course a pathological cell. (2) The immature form polymorphonuclear. In this there is a close resemblance to the neutrophile myelocyte but there is a nuclear indentation instead of the round nucleus of the myelocyte. It is this cell which often puzzles us as to whether to regard it as a true myelocyte. It is the meta-myelocyte of many authorities. (3) Between the mature or segmented polymorphonuclear and the immature one or metamyelocyte we have what may be designated the band-form nucleated one. These show the type of nucleus which one is familiar with in the nucleus of the transitional. (4) The mature, multilobed or segmented nucleus of the typical polymorphonuclear.

It would seem that if all tropical workers would agree upon some single method of recording differential counts it would be advantageous.

Under the blood findings in liver abscess, in a paragraph to follow in this chapter, I give suggestive counts indicating the value of Schilling-Torgau’s method.

In the differential count he not only divided up the polymorphonuclears but makes no separation of small from large lymphocytes. Although I have always divided lymphocytes into large and small ones I believe it unnecessary and unpractical and shall henceforth group all such cells in one grouping. The statement that large mononuclears and transitionals are cells of a similar origin, type and significance has always been my view.

SCHEME OF SCHILLING-TORGAU -------------------------------------------------+-------+--------------- | Normal| Percentage Type of Cell |Percen-|Moderate Sepsis | tage |(W. C. 14,000) -------------------------------------------------+-------+--------------- 1. Mast cells | 1 | 1.0 2. Eosinophiles | 3 | 1.5 { a. myelocytes | 0 | 0.5 3. Neutro- { b. immature forms (metamyelocytes) | 0 | 5.0 philes { c. band-form (Stabkernige) | 4 | 13.5 { d. multilobed (Segmentkernige) | 63 | 64.0 4. Lymphocytes | 23 | 10.5 5. Large mononuclears and transitionals | 6 | 4.0 -------------------------------------------------+-------+---------------

BLOOD CULTURING

Among tropical diseases, only malta fever, kala-azar and plague demand this method of diagnosis, although there are met commonly in the tropics many cosmopolitan diseases in which blood culturing is a principal diagnostic procedure. There are many ways of carrying out the cultivation of organisms from the blood but the one which may be strongly recommended is the following. The blood is obtained from a vein, the overlying skin of which has been painted with tincture of iodine to insure a sterile skin surface.

A stout hypodermic needle is attached to about 6 inches of rubber tubing which in turn is pushed over a downward bent glass tube which passes through a doubly perforated rubber stopper. A second glass tube, which also passes through the stopper, is bent upward to be attached to a second piece of rubber tubing for use in suction by the mouth. The glass tubes project about ½ inch below the under surface of the rubber stopper and above are about 2½ inches including the bent arm. This system of tubing and stopper is readily sterilized by boiling in a pan or instrument sterilizer. As a receptacle for the blood we employ Erlenmeyer flasks of 100 cc. capacity, containing 25 cc. of salt solution with 1% of sodium citrate, for prevention of coagulation. Blood that contains 0.2% of sodium citrate will not coagulate so that a 0.5% solution could be used instead of the usual 1% one. These citrated salt solution flasks are plugged with cotton, sterilized and kept on hand ready for immediate use, so that we only have to sterilize the stopper and tubing by boiling and flame the neck of the flask when removing the cotton plug to insert the stopper of the system. By suction we can take any amount of blood desired. I usually count the drops of blood as they fall into the citrated salt solution allowing 16 drops to the cc. In this way we may take from 10 to 25 cc. of blood at the bedside and then later on in the laboratory, when it is convenient, inoculate various media from the flask. For plates add 2 or 3 cc. of this citrated blood to 6 or 8 cc. of melted agar at 45°C. The blood mixture can also be added to various sugar bouillons for fermentation reactions. Finally we place the receiving flask in the incubator and culture it as well as the other media.

=Clot Cultures.=—A very simple method is to take blood with a Wright U-tube. Then centrifuge and use the serum for agglutination tests and the clot, emulsified in some liquid medium, for the blood culturing. For paratyphoid culturing _bile media_ are preferable, just as for typhoid.

=Lyon Blood Tube.=—Quite recently I have been using the blood tube recommended by Lyon. To make it, heat a 5- or 6-inch section of ¼ inch tubing in the centre and draw out as for making 2 bacteriological pipettes. Divide and seal off the large end in the flame. Next seal off the capillary end. Then apply a very small flame to a point on the large end just before it begins to taper to the capillary part. The heat causes the heated sealed-off air inside to force out a blow hole. To use: Break off the sealed capillary end and allow the capillary end to suck up blood from a drop just as with the Wright tube. I consider this tube superior to the Wright one.

=N. N. N. Medium.=—In culturing blood for protozoa the N. N. N. medium is usually employed. Novy and MacNeal originally used a 12½% meat infusion containing 2½% agar, 2% peptone, 1% normal sodium carbonate solution and ½% salt. To one part of this agar, melted and cooled to 60°C., they added twice the amount of defibrinated rabbit’s blood. In the N. N. N. medium, as modified by Nicolle, there is beside the blood only salt and agar—no peptone or meat extractives.

Citrated salt solution was the medium used by Rogers in the cultivation of splenic juice from kala-azar patients.

THE TAKING OF BLOOD FOR SEROLOGICAL TESTS

This can be done with the Wright tube, pipetting off the clear serum after centrifuging. We usually draw blood from a vein by use of the system of stopper and tubing described under blood culturing but employing an empty, sterile centrifuge tube.

Agglutination Tests

There are two methods of testing the agglutinating powers of a serum—the microscopical and the macroscopical or sedimentation method.

=For the microscopical method= draw up serum to the mark 0.5 of the white pipette. Then draw up salt solution to the mark 11. This when mixed gives a dilution of 1 to 20. One loopful of the diluted serum and one loopful of a bouillon culture or salt solution suspension of the organism to be tested gives a dilution of 1 to 40. One loopful of the 1-20 diluted serum and 3 loopfuls of the bacterial suspension give a dilution of 1-80. These two dilutions answer in ordinary diagnostic tests. The red pipette with a 1-100 or 1-200 dilution may be used where dilutions approaching 1-1000 are desired. Having mixed the diluted serum and the bacterial suspension on a cover-glass, we invert it over a vaselined concave slide and examine with a high power dry objective (⅙ inch). It is simpler to make a ring of vaseline to fit the cover-glass and make the mixture of diluted serum and culture in the centre of this ring or square. Then apply the cover-glass, press it down on the vaseline ring and examine as with the ordinary hanging drop. In making dilutions it is preferable to use salt solution, as the phenomenon of agglutination requires the presence of salts. Ordinarily, thirty minutes is a sufficient time to wait before reporting the absence of agglutination. Agglutination is more rapid at body temperature than at room temperature. In reporting agglutination, always give time and dilution. It is absolutely necessary that a control preparation be prepared in every instance; that is, one with the bacterial culture alone or with a normal serum of the same dilution as the lowest used. Some normal sera will agglutinate in 1 to 10 dilution, and group agglutinations (as paratyphoid with typhoid serum) may occur in 1 to 40 or possibly higher. It is very unusual for sera to agglutinate any other bacteria then the specific one in dilutions as high as 1-80.

=Macroscopic Agglutination.=—For the macroscopical or sedimentation test, take a series of small tubes (⅜ × 3 inches) and deposit 1 cc. of salt solution in each of the series. Now, having taken an empty test-tube, drop 4 drops of serum in it and then add 12 drops of salt solution. This approximately gives 1 cc. of a 1 to 4 dilution of the serum. It is more exact to make the 1 to 4 dilution with a graduated pipette. With a rubber-bulb capillary pipette, which has been graduated to hold 16 drops or 1 cc., draw up the contents of the tube containing the 1 to 4 serum and add it to the next tube containing 1 cc. of salt solution. This gives 2 cc. of a dilution of 1 to 8. Now mix thoroughly by drawing up and forcing out with the bulb pipette, and then withdraw 1 cc. and add to the next tube containing 1 cc. of salt solution. This gives a dilution of 1 to 16. Having mixed as before, again withdraw 1 cc. of the mixture and add it to the 1 cc. in the next tube. We now have a dilution of 1 to 32. Again withdrawing 1 cc. and adding it to the fourth tube containing 1 cc. of salt solution we have a dilution of 1 to 64. In tube 1 there is now 1 cc. of a dilution of the serum of 1 to 8; in tube 2, there is 1 cc. of a dilution of 1 to 16; in tube 3 of 1 to 32. Tube 4 contains 2 cc. of 1 to 64. The dilutions can be carried on in the same manner to any extent that may be desirable. In cholera agglutinations we may run up to 1 to 5000 or thereabouts. Of course, where such dilutions are employed, we generally start with 2 cc. of 1 to 50 in the first tube. When we have completed the series, each tube having 1 cc. of diluted serum, and the last 2 cc., we remove with the pipette 1 cc. from the last tube and discard it by ejection from the pipette leaving 1 cc. in the last tube. Now adding 1 cc. of a culture of typhoid or any other organism, we have the dilution of the serum in each tube doubled. Tube 1 now contains a serum in dilution of 1 to 16,

## acting on the bacteria; tube 2 of a 1 to 32; tube 3 of a 1 to 64.

Now place these tubes in the incubator and, after two to five hours or overnight, we examine for the clearing up of the supernatant fluid. If the serum in a certain dilution agglutinates, the clumps gravitate to the bottom and the upper part becomes clear. If so desired, these dilutions may be carried on to 1 to several hundred in the same way. It is safer to work with dead cultures instead of living ones. To prepare, take a twenty-four-hour agar slant culture of typhoid or paratyphoid and emulsify in salt solution (about 6 cc. to a slant).

By adding 0.1 of 1% of formalin to the typhoid emulsion and placing in the ice-box the cultures will be found sterile in about three days. The emulsion should be shaken twice daily while undergoing sterilization in the ice-box. Such cultures are not easily contaminated and appear to retain their agglutinable qualities for several months. The macroscopic methods are preferable with such dead cultures. For our Dreyer emulsions we use a two-billion suspension of typhoid or para-typhoid organisms in 1 cc. of the formalinized culture.

_Combination of Microscopical and Macroscopical Methods._—Microscopic: Prepare dilutions of serum as above described and take from each or several of the series, a loopful of the diluted serum. For control use a loopful of salt solution. Place on a cover-glass and add loopful of bouillon culture of the living organisms. Make hanging drop preparation, report after one hour at room temperature. Use ⅔ inch lens for examination.

Macroscopic: Add to each of the series, including the control, an equal amount of an emulsion of killed organisms.

The method of using a slide with two vaselined rings, one containing an emulsion in the specific serum and the other in salt solution, is of great practical value. This method is described under cholera.

_Complement Fixation._—Complement fixation tests have been employed in the diagnosis of several tropical diseases but do not seem to be at present sufficiently reliable or practical with the exception of that for yaws and tularaemia. The chief difficulty with complement fixation tests for suspected sera is to obtain a reliable antigen. Should we later on be able to prepare bacterial antigens as satisfactory as Noguchi’s acetone-insoluble antigen is for the Wassermann test there may be a field for such tests in tropical pathology.

OTHER PRACTICAL METHODS OF HAEMATOLOGICAL STUDY

Haemoglobin Estimation

The standard method now is the estimation of the oxygen capacity of the blood, using some gas apparatus, such as Van Slyke’s. Otherwise, the most accurate instrument for this purpose is the Miescher modification of the v. Fleischl haemoglobinometer.

[Illustration: FIG. 151.—Sahli’s Haemoglobinometer. (Greene.)]

The apparatus is expensive, requires considerable time and care in the making of estimations, and is exclusively an instrument for a well-equipped laboratory.

=Sahli’s Haemometer.=—A simple and apparently very scientific instrument which has been recently introduced is the Sahli modification of the Gower haemoglobinometer. Instead of the tinted glass, or gelatin colored with picrocarmine to resemble a definite blood dilution, Sahli uses as a standard the same coloring matter as is present in the tube containing the blood. By acting on blood with 10 times its volume of N/10 HCl, haematin hydrochlorate is produced, which gives a brownish yellow color. In the standard tube, which is sealed, a dilution representing 1% of normal blood is used. To apply this test, pour in N/10 HCl to the mark 10 on the scale of the graduated tube. Add to this 20 cubic millimeters of the blood to be examined, drawn up by the capillary pipette provided. So soon as the mixture assumes a clear bright dark-brown color, which requires about ten minutes, add water drop by drop until the color of the tubes matches. The reading of the height of the aqueous dilution on the scale gives the Hb. reading. The tubes are encased in a vulcanite frame with rectangular apertures. This gives the same optical impression as would planoparallel glass sides.

The most accurate readings are obtained with artificial light in a dark room but almost as satisfactory comparisons can be obtained with natural light from a window. It is advisable to turn the ruled side around so that one may match colors without being influenced in his determination by the scale.

The apparatus must be kept in a dark place as strong light will change the color of the standard tube. It is recommended that the N/10 HCl be preserved with chloroform.

The Dare instrument is excellent.

Pappenheim has recently proposed an instrument in which the blood is converted into haematin hydrochloride as for the Sahli apparatus. Instead of matching a standard tube, with a dilution made drop by drop in the second tube, the new method employs a wedge-shaped glass vessel showing graduations of the brown colored blood, the treated blood being matched against the wedge-shaped container (Autenreith-Koenigsberger Haemocolorimeter).

=Tallquist’s Haemoglobin Scale.=—This is a small book of specially prepared filter-paper with a color-scale plate of ten shades of blood colors. These are so tinted as to match blood taken up on a piece of the filter-paper and are graded from 10 to 100. So soon as the blood on the filter-paper has lost its humid gloss, the comparison should be made. This is best done by shifting the blood-stained piece of filter-paper suddenly from one to the other of the holes cut in each shade—the piece of filter-paper being underneath the color plate. At least a square centimeter of the filter-paper should be stained by the blood. Daylight coming from a window to the rear or at the side should be used in making the comparison. The error with this method is probably not over 10% after a little experience. If the colored plate is not kept in the dark, the tints tend to fade.

NORMAL BLOOD

In considering what may be termed normal blood, it must be borne in mind that the normal varies for men, women, and children:

Hb. Red Cells Leucocytes

Men, 90 to 110%, 5 to 5½ million, 7500. Women, 80 to 100%, 4½ to 5 million, 7500. Children, 70 to 80%, 4½ to 5 million, 9000.

COLOR INDEX

This is obtained by dividing the percentage of the haemoglobin by the percentage of red cells, 5,000,000 red cells being considered as 100%.

To obtain the percentage of red cells it is only necessary to multiply the two extreme figures to the left by two. Thus if a count showed the presence of 1,700,000 red cells the percentage would be 34 (17 × 2 = 34). If the Hb. percentage in this case were 50, then the color index would be 50 ÷ 34, or 1.4.

In normal blood the color index is, approximately, 1.

In anaemias we have three types of color index: 1. The pernicious anaemia type which is above 1. Here we have a greater reduction in red cells than we have of the haemoglobin content of each cell. For example, in a case of pernicious anaemia we have 2,000,000 red cells (40%) and 60% of haemoglobin, 60 ÷ 40 = 1.5. 2. The normal type, when both red cells and haemoglobin are proportionally decreased, as in anaemia fallowing haemorrhage. 3. The chlorotic type. Here there is a great decrease in haemoglobin percentage, but only a moderate decrease in the number of red cells. Hence the color index is only a fraction of 1. For example, in a case of chlorosis we have 40% of haemoglobin and 4,000,000 red cells, 40 ÷ 80 = 0.5.

One can judge fairly well the approximate color index by noting the character of the staining of the red cells. This is faint in bloods of low color index and deeper than normal in cells in a case with high color index.

TESTS FOR AGGLUTINATION AND HAEMOLYSIS OF THE RED CELLS (TRANSFUSION)

Transfusion of blood has become a method of greatest value in many types of anaemia.

In the selection of a donor for blood for transfusion it is always necessary to try his red cells against the serum of the recipient as well as the patient’s red cells against the serum of the donor, in order to prove the absence of haemolyzing or agglutinating bodies.

Certain persons have isohaemolysins in their blood which dissolve the red cells of other persons and in paroxysmal haemoglobinuria autohaemolysins may be present which can destroy the patient’s own red cells. This autohaemolysin seems operative only when a low temperature is followed by a high one. When haemoglobinaemia exists the liver converts it into bile pigment, causing bilious stools and jaundice. If one-sixth of the red cells are destroyed haemoglobinuria results.

In the following tables, two groupings of blood are given. Both are quoted in text-books, and both are in common use. Although that of Moss is more generally followed in France, England and the United States, the obvious desirability of having one classification universally employed, in order to avoid confusion and the possibility of serious accidents, has led to the recommendation that, on the basis of priority the grouping of Jansky be adopted.

In 1907, Jansky described the following four groups.

Group 1, the serum of which agglutinates the corpuscles of Groups 2, 3 and 4, while the cells are not agglutinated by any serum.

Group 2, the serum of which agglutinates the corpuscles of Groups 3 and 4, but not those of Groups 1 and 2, while the corpuscles are agglutinated by the serum of Groups 1 and 3, but not by those of Groups 2 and 4.

Group 3, the serum of which agglutinates the cells of Groups 2 and 4, but not those of Groups 1 and 3, while the corpuscles are agglutinated by the serum of Groups 1 and 2, but not by those of Groups 3 and 4.

Group 4, the serum of which does not agglutinate any corpuscles, while the corpuscles are agglutinated by the serum of all other groups.

In 1910, Moss made the following classification:

Group 1, the serum of which does not agglutinate any corpuscles, while the corpuscles are agglutinated by the serum of Groups 2, 3 and 4.

Group 2, the serum of which agglutinates the corpuscles of Groups 1 and 3, while the corpuscles are agglutinated by the serum of Groups 3 and 4.

Group 3, the serum of which agglutinates the corpuscles of Groups 1 and 2, while the corpuscles are agglutinated by the serum of Groups 2 and 4.

Group 4, the serum of which agglutinates the corpuscles of Groups 1, 2 and 3 while the corpuscles are not agglutinated by any serum.

At the present time it is accepted that the four groups considered include all adult persons; i.e., that the classification is complete.

=Before transfusing= carry out the following tests:

From a vein take about 1 cc. of blood in a centrifuge tube containing 1% of sod. citrate salt solution; then shift the stopper of the blood system to a dry centrifuge tube and draw into it about 3 or 4 cc. of blood. Throw down the citrated blood, pipette off the supernatant fluid and wash the sediment with normal saline.

Again pipette off the saline after centrifuging and make a 10% emulsion of the red-cell sediment in normal saline.

Centrifuge the coagulated blood in the other tube and collect the serum which separates from the clot.

Carry out these procedures for both donor and recipient.

Tests: 1. In a small test-tube deposit 1 drop of the donor’s 10% red-cell emulsion and then add 4 drops of the recipient’s serum.

2. Treat similarly 1 drop of the recipient’s red-cell emulsion with 4 drops of the donor’s serum.

3. Treat 1 drop of donor’s red-cell emulsion with 4 drops of his serum.

4. Treat 1 drop of recipient’s red-cell emulsion with 4 drops of his serum. Finally add 1 cc. of salt solution to each of the four tubes, shake gently and place in incubator for two hours.

5. Treat one drop of donor’s red-cell emulsion with four drops of salt solution.

6. Treat one drop of recipient’s red-cell emulsion with four drops of salt solution.

Tubes 5 and 6 are controls of saline.

Tests 3 and 4 should fail to show either agglutination or haemolysis. If agglutination or haemolysis appears in tubes 1 or 2, the donor is not satisfactory; but if agglutination appears in tube 2 only, he may be used in an emergency.

Some prefer to keep the tubes overnight in ice-box after the preliminary examination following incubation.

_Lee’s Technique._—For the regular carrying out of this method one should keep on hand the sera of individuals belonging to groups 2 and 3 (Moss). To carry out the tests prepare a suspension of the donor’s red cells by dropping 2 or 3 drops of his blood into 1 cc. of citrated salt solution. Deposit a platinum loopful of standard serum 2 on a slide and emulsify in it a loopful of the donor’s red-cell suspension. A concave slide with two concavities is convenient, the serum-cell emulsion being made on the cover-glasses which are to be inverted over the vaseline ringed concavities. The agglutination can be observed with a high power magnifying glass or the ⅔-inch objective. Agglutination, when it occurs, is usually complete in five to fifteen minutes. Repeat test with serum 3. If both test sera agglutinate the donor’s cells he belongs to group one. If neither agglutinate, to group four.

Agglutination by group two serum but not by three puts the donor in group three. Agglutination by group three serum but not by group two shows a group two donor. It would seem safe to use the cells of any donor of group 4, as such cells are not agglutinated by the sera of any group. It is, however, advisable to try to obtain a donor whose blood belongs to the same group as the donee. When standard sera 2 and 3 are not on hand one may use the following _emergency method_ of Lee:

“A small amount of blood is collected from a patient (1 cc. from the ear or finger is sufficient), and allowed to clot. The serum is then obtained. One drop of this serum is placed on a slide and mixed with a drop of suspension of blood of the donor taken into 1.5% citrate solution. (A few drops of blood are taken into approximately 10 times the amount of 1.5 citrate solution and shaken. It is very important that the blood be dropped directly into the citrate, and should not be partially coagulated.) The test will appear in a few moments, and is best examined under the microscope, where, in the event of a positive test, marked agglutination will be evident. The test will also be evident macroscopically. In the event of a negative test it is a wise precaution to raise the cover-glass, and after making sure that the serum and cells are well mixed, to examine the preparation again. The only possible source of confusion is the appearance of rouleaux of the red corpuscle, indicating a too thick emulsion. If the test is negative, transfusion may be regarded as entirely safe.”

In the absence of agglutination haemolysis never occurs. Only about one-fifth of agglutinating sera prove also haemolytic. Rarely a pernicious anaemia patient’s serum may agglutinate his own red cells. This auto-agglutination is regarded as an important test in acquired haemolytic jaundice.

OCCULT BLOOD

When the presence of blood in the faeces, gastric contents, urine or body fluids, is suspected but cannot be recognized by macroscopic or microscopic methods, it is necessary to resort to spectroscopic or chemical tests. These tests are, however, individually unsatisfactory. The spectroscopic method is not delicate, the haemin-crystal method does not give uniform results and the various color tests, although very sensitive, are given by many substances other than blood. Consequently, it may be said that, with the color tests, it is negative results that are significant, and with other than the color tests it is positive findings that are informative. Serological tests are the most satisfactory medicolegally.

_Haemin Crystal Test (Teichmann)._—Prepare a solution (stable) of 0.1 gm. each of KI, KBr, and KCl in 100 cc. acetic acid. Mix a few drops with some of the material on a slide, apply a cover-glass, and _gently_ warm until bubbles begin to appear. Then cool _slowly_, and examine for the characteristic dark-brown crystals.

_Haemochromogen Crystals (Donogány)._—Mix one drop each of suspected fluid, pyridin, and 20% NaOH on slide, and let dry. If positive, radiating needles will form after several hours.

_Spectroscopic Tests._—These depend upon the recognition of the characteristic absorption spectra of haemoglobin or its derivatives (Fig. 24). The degree of concentration influences their appearance, and one should start with a relatively concentrated solution, diluting cautiously until the bands are typical.

The small, direct-vision (hand) spectroscope suffices. A wavelength scale is a convenient attachment. Daylight or strong artificial light (such as the “daylite” lamp) is used. Have solution in a small test tube or, preferably, a flat cell with a thickness of about 1 cm. Before use, focus Frauenhofer’s lines sharply.

Reducing agents are employed, such as ammonium sulphide, or Stokes’ solution made up as follows: Dissolve 3 gm. FeSO_{4} in cold H_{2}O; add cold, aqueous solution of 2 gm. tartaric acid; make up to 100 cc.; immediately before use, add strong NH_{4}OH until precipitate first formed is dissolved. Both solutions must be freshly prepared, and the sulphide must be warmed to about 50°C.

Material that is uncontaminated, relatively fresh and in relatively concentrated aqueous solution may give any or all of the upper three spectra, a few drops of reducer changing the first to the second.

If the material is older, dissolve the suspected stain in 1-2 cc. of 10% NaOH, heat almost to boiling, cool, and add a few drops of reducer. Examination shows Spectrum 5.

It is better, however, especially with much contamination, to prepare an ethereal, acid extract. After having ground the material thoroughly with water, if it is not already in liquid form, shake it with an equal volume of neutral ether. Reject ether extract, and, to 10 cc. of residue, add 3 to 5 cc. of glacial acetic acid. Shake thoroughly with an equal volume of ether. If the ether does not separate readily, mix gently with a few drops of alcohol. Remove ethereal extract, and evaporate it to a small bulk for use in tests. Examination will show spectrum of acid haematin, which, however, in ethereal solution, resembles Spectrum 3 more than 4.

_Donogány’s Method_ increases the delicacy of the spectroscopic test, and is also a color test. Dissolve the pigment with 20% NaOH, add fresh pyridin and, if necessary, fresh ammonium sulphide. Filter. The filtrate will be more or less orange-red according to blood content, and will show Spectrum 5.

_Color Tests._—The reliability of these may be enhanced by the use of methods which involve the removal or destruction of interfering substances. In such a method, the original aqueous solution is boiled for 15 to 20 seconds, and the acid ethereal extract is prepared as previously described. This extract is dropped on filter-paper, the reagents being applied to the moistened spot. The delicacy of these several tests is variable, being greater with blood in aqueous solution than in biological fluids, but it may be given as approximately 1-25,000 for the guaiac and aloin, and 1-250,000 for the benzidine test.

(_a_) Treat moist spot with a few drops of freshly prepared 2% alcoholic solution of _guaiac_ resin, and then a few drops of hydrogen peroxide. A blue color is “positive.”

(_b_) Treat moist spot with a few drops of 3% _aloin_ in 70% alcohol, and then with ozonized turpentine (turpentine that has stood for a few days in an open vessel in sunlight). A purplish-red color within 10 minutes is “positive.”

(_c_) Treat moist spot with 2 drops of glacial acetic acid, a few crystals of _benzidine_ (preferably white), and finally 2 drops of hydrogen peroxide. A greenish-blue color is “positive.”

ACIDOSIS

Everyone is familiar with that form of respiratory disturbance associated with diabetic coma that is known as Kussmaul’s air hunger. Here we have hyperpnoea, a form of dyspnoea typically without cyanosis, and furnishing the best clinical evidence of acidosis. Acidosis, however, is now recognized to be but a particular phase of disturbance of the _acid-base equilibrium_ of the body, and recent work has radically changed our conceptions of its features and its intricate relationships.

Van Slyke restricts the use of the term “acidosis” to describe a condition caused by acid retention sufficient to lower either the bicarbonate or the pH of the blood below normal limits. The pH of the blood may be considered the danger sentinel; as long as it is normal, the acid-base equilibrium is normal or compensated; otherwise, it is uncompensated, and life is seriously threatened. The normal pH of the blood may be given as 7.3 to 7.5 (a slightly alkaline reaction), each individual, however, probably having normally narrower limits of variation. That of the blood serum is about 0.2 pH higher, and that of the other body fluids (not the excretions) probably closely approximates and promptly follows any change in that of the blood plasma. Variations to the acid side may, for a short time at least, be as low as 7.0, although not much lower without fatal results; 7.0 is considered the point where coma occurs. Variations to the alkaline side (Alkalosis) beyond 7.8 are accompanied by symptoms of tetany, although one is not at present justified in assuming that all tetany is either caused, or accompanied, by alkalosis. So, the extreme range of reaction compatible with life probably lies approximately between pH of 7.0 and 7.8.

Recent, but as yet unconfirmed, work suggests that the severe reactions following intravenous medication or infusions may be due, at least in part, to the fact that the pH of the fluid introduced is decidedly more acid or alkaline than that of the blood. This applies to solutions of glucose, the salines, and possibly also to sodium citrate, arsphenamine, sera, antitoxins, etc. The question of suitably buffering such solutions, _e.g._, with suitable phosphate mixtures, in order to avoid disturbance of the acid-base equilibrium, is being studied, and the preliminary results are promising.

The hydrogen-ion concentration (or its derivative, pH) of the blood varies as the ratio between the concentrations of dissolved carbonic acid and bicarbonate (generally indicated by (H_{2}CO_{3})/(NaHCO_{3})), i.e., a relative increase in the H_{2}CO_{3} increases the hydrogen-ion concentration and lowers the pH, and vice versa. The stability of this ratio is preserved by body mechanisms operative in controlling its two factors,—the H_{2}CO_{3} being under respiratory control, and the NaHCO_{3}, considered as representing the alkali reserve, being normally maintained by food.

The erythrocytes control the concentration of bicarbonate by virtue of their haemoglobin and the reversible reaction.

H_{2}CO_{3} + NaCl ⮂ HCl + NaHCO_{3}

The HCl passes into the cell, and is probably held by the haemoglobin. In the lungs, the CO_{2} is excreted and NaCl reformed. This ability of haemoglobin to form bicarbonate is important inasmuch as the corpuscles can conceal 5 to 10 times as much acid as the plasma bicarbonate can ordinarily neutralize. A full appreciation of the significance of this ratio being the basis of an intelligent comprehension of acid-base equilibrium, a detailed analysis of factors that tend to influence the ratio is given.

Factors Operating:

_A. To increase or protect bicarbonates:_ 1. Administration of bicarbonate. 2. Loss of gastric HCl induced by obstructing the pylorus, and regularly washing out the stomach for some days. 3. Processes indicated by increased excretion in the urine of ammonia compounds, the ammonia being probably diverted from urea formation, and of substances producing a titrable acidity, they including buffer acids such as acid phosphates. 4. Possibly a shift of HCl to the tissue cells from the plasma like that from the plasma to blood cells.

_B. To decrease bicarbonate:_ 5. Acid substances, by their (_a_) Increased production. (_b_) Decreased elimination or (_c_) Ingestion. 6. Diuresis, with elimination via the urine. 7. Lack of Factor A (3). 8. The hyperpnoea associated with deficient oxygen.

_C. To increase carbonic acid:_ 9. Administration of carbonic acid. 10. Impaired diffusion in the alveoli of the lungs. 11. Slowing of respiration.

_D. To decrease carbonic acid:_ 12. Hyperpnoea. (_a_) Voluntary. (_b_) Due to disease processes. (_c_) Due to low oxygen content of air. (_d_) Emergence from warm water. 13. Low atmospheric content of CO_{2}.

[Illustration: FIG. 152.—Carbon dioxide absorption curves. (Modified from Peters, Barr, and Rule, and Van Slyke).]

Figure 152 is a graphic representation of essential facts in acid-base equilibrium. Ordinates represent total CO_{2} content, which comprises that in simple solution and that as bicarbonate of whole blood in volumes per cent, and abscissae the mm. CO_{2} tension in the blood as drawn. The line OT gives the proportion of total CO_{2} present in simple solution. pH values are shown by the lines OL, OM, etc. The extreme normals for carbon dioxide absorption curves are OP and OR. The CO_{2} tension of alveolar air may be the same or vary as much as 20 mm. below, while that of venous blood will be about 6 (0.8-10.0) mm. higher than that of arterial blood. The “CO_{2} capacity” (or “CO_{2} combining power”) of plasma may be as much as 15 vol. % more than the total CO_{2} of whole blood.

The actual state of acid-base balance, then, can only be determined by the use of any two of a number of interdependent variables, such as total CO_{2}, CO_{2} tension, pH, H_{2}CO_{3} concentration, other buffers than bicarbonate, plasma chloride, ratio of oxyhaemoglobin to haemoglobin, etc. Findings that fall within ABCD and at about 40 mm. tension indicate a normal equilibrium for the resting individual at ordinary altitudes. Or, such a normal would be a total CO_{2} of about 49 (43-56) vol. % for whole blood, and 50-65 vol. % for plasma. The normal for the individual falls within narrower limits.

If either H_{2}CO_{3} or bicarbonate varies from normal values, there is apparently an effort on the part of the body to compensate by adjusting the other so as at least to maintain a normal pH. This is accomplished by respiration, or by diverting alkali from or recalling it to the blood stream. Naturally, treatment of any such abnormal condition will do well to imitate Nature’s efforts. Haggard and Henderson have demonstrated that blood alkali may be decreased in two ways—by acids (the _acidotic process_) or by acapnia (the _acapnial process_). By the forced breathing of the acapnial process the lungs are over-ventilated and an excessive amount of carbon dioxide is washed out of the blood, thus bringing on a temporary alkalosis. In case of prolonged forced breathing, nature prevents an extreme alkalosis by causing the alkali to leave the blood, it being stored in the tissues or excreted in the urine. Blood relatively poor in carbonic acid or relatively rich in alkali acts to depress respiration, and the slowing of respiration produces an acidosis by the resultant retention of H_{2}CO_{3}, this causing alkali to be recalled from the tissues. Thus acidosis, in calling more alkali into the blood from the tissues, represents what may be regarded as a restorative effort. Hence, administration of bicarbonate is indicated in acidotic processes, and of CO_{2} in acapnial; the use of the wrong one is dangerous.

The numbered regions of the chart are associated with various clinical conditions, e.g., tetany from 1, 2, and 3; the acidosis of diabetes mellitus, nephritis, or infantile marasmus with 6 or 9; pneumonia, morphine narcosis, and breathing of air containing 3-5% CO_{2} with 7 or 8; emphysema with 4; some cardiac cases with 9; overdose of bicarbonate with 1 or 4; fever with 2; as the result of high altitudes, 2 or 3, or, when acclimated, 6; shock (handling of intestines), deep ether anesthesia, and carbon monoxide asphyxia with lowered bicarbonate. The disturbances of acid-base equilibrium in the last two are the result of acapnial processes.

Normal metabolism results in the constant formation of acids, especially H_{2}CO_{3}, and disease processes may occasion the presence of still more. A constant loss of alkali results, the neutralization products being eliminated mostly in the urine, and the H_{2}CO_{3} via the lungs; the body fluids are excellently buffered, the most important buffers being bicarbonate, proteins (especially haemoglobin), and phosphates. In the maintainance of the normal pH, the CO_{2} (or H_{2}CO_{3}) is the easily variable factor. The onslaught of invading acids is first met by the bicarbonates (acidotic process); hyperpnoea lowers the H_{2}CO_{3} and a normal pH is maintained until the bicarbonates are reduced to one-fourth (perhaps even to one-eighth) of their normal concentration. If, nevertheless, the pH falls (and only then), the other buffers are used, and, if it reaches 7.0, most of the remaining bicarbonate becomes available. The blood handles the situation, but buffers from the tissues or other body fluids also become available in extreme cases.

As noted above, measurement of one variable will be inadequate exactly to determine the state of acid-base equilibrium. As long, however, as the pH is normal, which is the usual finding in most pathological conditions, including mild acidosis, one determination will suffice. Clinical methods comprise tests for whole blood or plasma CO_{2} or bicarbonate, alveolar CO_{2} tension, bicarbonate tolerance, pH of blood or urine, Sellard’s test, NH_{3} quotient of urine, or presence of abnormal acids (particularly acetone bodies) in blood or urine. The first two methods are the ones of choice,

## particularly the first, as, by it, one can estimate the reserve

of the very important blood buffer, bicarbonate, and its result closely indicates the total buffers.

The _tolerance for bicarbonate_ is a very convenient and practical measure of acidosis, and means the dose of NaHCO_{3} required to produce a urine alkaline or amphoteric to litmus. A normal finding is 5-10 grams; 20 is required with a mild, 30-40 with a more severe, and more than 40 gm. with extreme degrees of acidosis. In coma, it is usually impossible to produce an alkaline urine.

Certain changes in the urine are recognized and acceptable as indirect evidence of acidosis, but these changes are not synonymous with acidosis, being dependent in part upon renal integrity and other factors. The NH_{3} quotient of urine (ammonia nitrogen; total nitrogen), as usually determined with patient on a mixed diet, is normally about 5%. Values of 10-40% occur in acidosis. It may be increased by diet, disturbances of protein metabolism, ammoniacal fermentation, etc., and there may be no increase in certain diseases with acidosis. The ammonia probably does protect the blood alkali, but its efficacy is intimately associated with renal function, inasmuch as Nash and Benedict have presented strong evidence to the effect that urinary and blood NH_{3} is the product of an active synthetic function of the kidneys themselves. Acetone bodies in the urine (_ketonuria_), in the blood (_ketosis_), or in the breath, have diagnostic value but are poor indices of severity of acidosis and may be absent in acidosis. Acetone and diacetic acid have the same significance: a progressive increase gives a grave prognosis, and it is generally considered that the presence of β-hydroxybutyric acid indicates greater severity.

Acidotic acidosis is due either to the abnormal formation or ingestion of acid substances, or to decreased elimination of normal metabolic products. Ketosis is the important example of the former, and retention of acid phosphates of the latter. In either case, the body is robbed of its bases. The acidosis of diabetes mellitus is characterized by ketosis and increased NH_{3} quotient of urine, while that of nephritis is a phosphate retention without ketosis, and, as one would expect, the NH_{3} quotient is usually not increased. Infantile diarrhoea with ileocolitis shows a marked ketosis, but, lacking the ileocolitis, the ketosis is only moderate and the acidosis is due to phosphate retention.

The appearance of an acidosis in disease constitutes a serious development demanding immediate attention. It is usually present at time of death and may be the immediate cause. We must be prepared for the appearance of acidosis in the course of numerous cosmopolitan diseases, and its presence has been recognized in a few tropical conditions. Before we generally recognized the great importance of the acidosis factor in pathology, there were two standard treatments for _yellow fever_ and _blackwater fever_, the Sternberg one in the former and the Hearsey one for the latter, both of which had as a basis the administration of alkalis, which is our best means for neutralizing the deleterious action of increased acid production in the body or defective elimination of the same. It was a very important contribution to the therapeutics of _cholera_ when Sellards, recognizing the tendency of the nephritis to produce an acidosis in this disease, made use of intravenous injections of NaHCO_{3} to combat the condition, thus counteracting the anuria, one of the chief complications leading to death. More recently, the Egyptian workers noted an acidosis in _kala-azar_, a finding verified and emphasized by Rogers. There is also an acidosis in _heat stroke_, so that intravenous or rectal injections of NaHCO_{3} are of value. It will thus be seen that acidosis is a most important condition to keep in mind in tropical conditions, and it will be well to be on the watch for other varieties of disturbance of acid-base equilibrium.

Relative to the _administration of bicarbonate_ in treatment, there is now a decided reaction against the use of amounts that may prove injurious by reason of the danger of alkalosis. There is a tendency to employ it only in decompensated acidosis, and control it by estimations of plasma CO_{2} capacity, 0.5 gm. NaHCO_{3} per 19 kg. body weight will raise the plasma CO_{2} capacity by 1 vol %. It is distinctly contraindicated in cases whose low plasma CO_{2} is due to acapnial processes. Early administration is desirable in children, and good results are obtained, especially with the older ones. An acidosis, however, once established in infants may cause death despite alkali. In order to avoid over-dosage of bicarbonate, methyl red, which is more sensitive than litmus to early changes in the reaction of the urine, should be employed as an indicator. The appearance of a yellow color upon its addition to the urine is the sign to suspend further administration of alkali.

Glucose is indicated in conditions with ketosis due to carbohydrate deficiency, providing the organism can assimilate it.

CHEMICAL ANALYSIS OF BLOOD

The chemical analysis of the blood has attained a clinical simplicity and significance that demands recognition. It provides points of value in diagnosis, prognosis, and treatment, being especially useful in nephritis, diabetes, acidosis, comatose conditions, gout, and in questions of renal function and treatment, especially dietetic. Urine findings are always dependent upon kidney function, and, by blood chemistry, we can pass behind this barrier.

Few diseases have been as yet studied thoroughly in this respect, but our fund of knowledge is receiving constant additions. The field of tropical medicine is practically untouched, and it is quite possible that an investigation along this line might there yield facts of interest and value.

The following table (amplified from Myers), is a concise summary of normal findings and those encountered in various clinical conditions. The diagnostic significance is evident. Some of the results are based upon the analysis of many cases; others upon but few. One might include the findings mentioned elsewhere regarding acidosis in certain tropical diseases, but, except for such, we have no other data relative to them, unless one mentions that blood sugar is increased in the tropics. The values are given in milligrams per 100 cc. whole blood (the usual system), except those for diastatic

## activity (recorded in Winslow’s empirical units) and acidosis

(expressed in terms of plasma carbon dioxide combining power—volumes %). “Inc.” and “Dec.” signify increased and decreased respectively.

RESULT OF CHEMICAL EXAMINATION OF BLOOD

CONDITION --+-------+-------+------+-------+--------+-------+-------+-------+------ | Non | | | | | | | |Plasma |protein| Urea | Uric |Creati-| Sugar |Choles-|Chlor- | Dias- |CO_{2} | nitro-|nitro- | acid | nine | | terin | ides | tase |capa- | gen | gen | | | | | | | city --+-------+-------+------+-------+--------+-------+-------+-------+------ NORMAL | 25-300| 10-15 | 2-3 | 1-2 | 90-120 |170-250|450-500| 8-64 | 53-77 | | | | | | | | | DIABETES MELLITUS, MILD | | | | |150-300 | | | Inc. | | | | | | | | | | DIABETES MELLITUS, SEVERE | | 20 | 4-10 | 2-4 |300-1200| Inc. | Dec. | Inc. | 10-50 | | | | | | | | | NEPHRITIS, ACUTE | | 40-100| 5-15 | 2-6 |120-180 | | Inc. | 20-45 | | | | | | | | | | NEPHRITIS, INTERSTITIAL, EARLY | | 15-25 | 5-12 | 2-3.5 |120-150 | | | Inc. | | | | | | | | | | NEPHRITIS, INTERSTITIAL, TERMINAL |100-300| 60-300| 5-27 | 5-28 |120-240 | Inc. | Vari- | Inc. | 12-40 | | | | | | | able | | NEPHRITIS, PARENCHYMATOUS (NEPHROSIS) | 20-50 | 2-5 | 2-4 |120-200| Inc. | Inc. | | | | | | | | | | | | NEPHRITIS, CHRONIC DIFFUSE, SEVERE | | to 230| to 10| to 16 | to 250 | | | | | | | | | | | | | URAEMIA | 90-350| 70-300| | | | | | | | | | | | | | | | KIDNEY POLYCYSTIC, DOUBLE | | to 75| to 5 | to 8 | to 200 | | | | | | | | | | | | | PROSTATIC OBSTRUCTION | Inc. | 12-40 | 3-9 |1.5-3.5|110-160 | | | | | | | | | | | | | GOUT | | | 4-10 | | | | | | | | | | | | | | | HYPERTHY- ROIDISM | | Inc. | | | Inc. | | | Inc. | | | | | | | | | | HYPOENDOCRINE CONDITIONS | | Dec. | | | 60-90 | | Dec. | Dec. | | | | | | | | | | ECLAMPSIA | 25-45 | 10-25 | 4-8 | | | | | | 43-58 | | | | | | | | | INTESTINAL OBSTRUCTION, ACUTE | 75-170| 45-120| Inc. | Inc. | | | | | | | | | | | | | | FEVER, ACUTE | Inc. | Inc. | to 4 | | Dec. | Dec. | | | | | | | | | | | | PNEUMONIA, SEVERE AND LATE | | to 53 | to 18| to 3.5| to 180 | Inc. | Dec. | | Dec. | | | | | | | | | ANAEMIA, PERNICIOUS | to 108| to 75 | to 10| to 3.1| to 300 | Dec. | Inc. | | Dec. | | | | | | | | | MALIGNANCY, LATE | Inc. | Inc. | Inc. | Inc. | | Dec. | Inc. | | Dec. | | | | | | | | | DEMENTIA PRAECOX, CATATONIC | | 6-10 | Dec. | | Inc. | | | | | | | | | | | | | SHOCK | Inc. | Inc. | | Inc. | Inc. | | | | Dec. | | | | | | | | | BICHLORIDE OF MERCURY POISONING |to 370 | to 300| to 15| to 33 |120-200 | Inc. | | | | | | | | | | | | PLUMBISM | Inc. | Inc. | Inc. | | | | | | --+-------+-------+------+-------+--------+-------+-------+-------+------

_Interstitial nephritis_ is characterized by a nitrogen retention, while _parenchymatous nephritis_ has relatively little nitrogen retention but does have a decided tendency towards chloride retention. _Essential hypertonia_ with its normal blood chemistry is differentiated from _arteriosclerosis_ with its frequent nitrogen retention. The imminence of _uraemia_ may be judged by the extent of the nitrogen retention. We have an aid in the differentiation of the uraemia of nephritis accompanied by a flagging heart from the passive congestion of cardiac decompensation, especially as to which is the secondary condition, and thus therapeutic indications relative to mooted questions of treatment (hot packs, morphine, renal stimulants, etc.). Unsuspected cases of nephritis showing only gastric symptoms clinically have been detected by blood chemistry. The significance of albumin in traces and occasional casts in urine has been more definitely established by examination for increase of uric acid in the blood—an increase arguing for an organic lesion. Values of over 4 for creatinine do not occur without great impairment of renal function, and findings of more than 5 have practically uniformly foretold a fatal termination in less than six months, except in acute nephritis and mild bichloride of mercury poisoning. The creatinine is also the best guide to the status of renal function in terminal cases. The chloride and nitrogen content afford guides to diet.

The blood may indicate a prediabetic state, and place the practitioner upon his guard. There is a condition but recently recognized in which there is a normal blood sugar, a persistent glycuresis of usually less than 1% and independent of carbohydrate intake, occasionally polyuria, but with no other symptoms of diabetes mellitus. It is known as _renal diabetes_, is apparently harmless, probably not uncommon, and may represent the condition affecting most of those “diabetics” who can disregard diet with impunity. The blood sugar and plasma CO_{2} are usually considered the only safe guides in the treatment of _diabetes mellitus_ and no extended medical treatment or surgical interference should ever be attempted without their estimation. Glycosuria is a poor guide, especially in advanced cases.

In _comatose conditions_, nitrogen retention will indicate the uraemic, and hyperglycaemia the diabetic cases. But _acute nephritis_ should always be borne in mind, as it may have a pronounced acidosis but no nitrogen retention.

A high uric acid finding alone is characteristic of gout, and aids in differential diagnosis from simple rheumatic fever and other arthritides, any uric acid retention in them being usually accompanied by retention of other nitrogenous elements. It is especially useful in the diagnosis of gouty arthritis without tophi.

The efficacy of treatment will, in general, be shown by the degree of approach to normal blood findings.

LEUCOPENIA

This is a term used to designate a reduction in the normal number of leucocytes. A leucocyte count of 5000 would represent a slight leucopenia; one of 2000, a marked leucopenia. In the later stages of typhoid, and in acute miliary tuberculosis, we expect a moderate leucopenia. Glandular tuberculosis may give a very marked leucopenia. Tuberculous peritonitis will show moderate leucopenia or a normal count.

The leucopenia of typhoid is moderate and is often preceded in the first few days by a moderate neutrophile leucocytosis. Later on we have a decided increase in the lymphocytes. A marked diminution or absence of eosinophiles is so characteristic that any increase in eosinophilic percentage negatives a diagnosis of typhoid.

Paratyphoid gives a similar blood picture.

Chronic alcoholism and chronic arsenic poisoning cause a reduction in the number of the white cells. Pernicious anaemia, especially the aplastic type, shows a marked leucopenia, as is also the case with Banti’s disease. Two tropical diseases, kala-azar and dengue, show a marked leucopenia, the counts often being below 2500. During the apyrexial period of malaria we may have a white count of 5000.

It has recently been claimed that a leucopenia with a coincident marked reduction in the lymphocytes is characteristic of measles and that this occurs several days before the Koplik spots appear.

Kocher notes that in exophthalmic goiter the leucocyte count is considerably diminished and that the polymorphonuclears are not much more than one-half the usual percentage while the percentage of the lymphocytes is almost double the normal.

X-ray treatment tends to destroy leucocytes in the exposed region, especially polymorphonuclears. The small lymphocytes are least affected.

EOSINOPHILIA

Where the eosinophiles are increased to 5%, we have a moderate eosinophilia. In some cases of infection with intestinal parasites, especially hookworms, but also from other parasites, as round and whip-worms, we may have an eosinophilia of 30 to 50%. In Guam, among the natives, it is difficult to find an eosinophile count under 15%. The eosinophilia tends to disappear when the anaemia becomes very severe.

_Echinococcus_ infection has an eosinophilia which disappears when the cyst is removed. Continuance of the eosinophilia indicates that all cysts were not gotten rid of.

The eosinophilia of trichinosis is best known, and a combination of this blood finding with fever and marked pains of muscles, would justify the excision of a piece of muscle for examination for encysted embryos.

In true asthma eosinophilia is marked, and its absence is of value in indicating other causes for the condition. Certain skin diseases, especially pemphigus, show eosinophilia. Blastomycoses are usually found to show eosinophile increase.

An increase of eosinophiles always attracts attention to the possibility of intestinal parasite infections or to skin affections. The explanation of eosinophilia is obscure although Neisser regards the increased production of eosinophiles as an expression of sympathetic system irritation.

Eczema and psoriasis are not apt to give more than 3 or 4% eosinophiles. A rather high degree of eosinophilia is found in mycosis fungoides.

Scabies also gives an eosinophilia.

The proportion of eosinophiles in the blood of children is greater than in that of adults.

Increase of both eosinophiles and mast cells is found in myelogenous leucaemia.

An eosinophilia tends to appear following splenectomy. With a Wright stain showing acid tendencies one may count polymorphonuclears as eosinophiles unless noting smaller size of granules.

LEUCOCYTOSIS

It is to an increase in the polymorphonuclears that this term is usually applied, the term lymphocytosis or eosinophilia being employed where white cells of eosinophile or lymphocyte nature are increased. We have physiological leucocytosis in the latter weeks of pregnancy, also in the new-born, and in connection with digestion.

=Pathological Leucocytosis.=—Pneumonia. In this disease we have a leucocytosis of 20,000 to 30,000 or higher. The eosinophiles are almost absent. A normal leucocyte count in pneumonia makes a prognosis unfavorable.

The leucocyte count drops about the time of the crisis, and with the reappearance of eosinophiles is a favorable sign.

Toxaemic conditions as uraemia, diabetic coma and poisoning by CO_{2} tend to show a leucocytosis.

Septic processes. The leucocyte count is of great value, especially when we obtain a leucocytosis with 80 to 90% of polymorphonuclears, as in appendicitis, cholecystitis, or other suppurative conditions. A marked leucocytosis is of diagnostic importance in acute ulcerative endocarditis provided it is not fulminant in type.

According to Cabot, leucocytosis varies in infections as follows:

1. Severe infection—good resistance; early, marked and persistent leucocytosis.

2. Slight infection—slight resistance; leucocytosis present, but not marked.

3. In fulminating infections we may have no increase in whites, but a higher percentage of polymorphonuclears.

4. Slight infection and good resistance may not be productive of leucocytosis.

It is in connection with the question of operation in appendicitis or similar conditions that the matter of a leucocyte count is of prime importance. If there be a leucocytosis but with less than 75% of polymorphonuclears it indicates an infection of little virulence or a walled-off process with an exacerbation. It is difficult to form an opinion when the polymorphonuclears are under 80%. Leucocytosis with polymorphonuclear percentage of 85 to 90 indicates immediate operation; percentages over 90 point to peritonitis and if with such percentages of polymorphonuclears there is absence of leucocytosis the prognosis is grave.

The blood of cases with malignant tumors tends to show a moderate leucocytosis except in epithelioma of the skin. When a cancer is ulcerating quite a high white count may be obtained.

Spirochaete fevers, as relapsing fever, may give a leucocytosis of from 25,000 to 50,000.

Smallpox, especially at time of pustulation, plague, scarlet fever, and liver abscess give a leucocytosis of from 12,000 to 15,000.

Smallpox often shows a very large percentage of very characteristic large mononuclears.

The leucopenia and lymphocyte increase in measles are important points in differentiating it from scarlatina.

Influenza shows a leucopenia at first, then a leucocytosis and, following the fall in fever, a second lowering. The very fatal pneumonias of the 1918 epidemic of influenza showed a marked leucopenia.

With meningitis counts of 25,000 are not unusual, in abscess of the brain the white count rarely exceeds 15,000.

Poliomyelitis and polioencephalitis give a slight leucocytosis during the febrile accession.

Erysipelas and epidemic cerebro-spinal meningitis also give a leucocytosis of from 15,000 to 20,000. In malignant diseases we sometimes have a moderate leucocytosis. Rogers states that in liver abscess, with a leucocytosis of 15,000 to 20,000 we have only about 75 to 77% of polymorphonuclears—there being also a moderate increase in the percentage of large mononuclears.

Drugs such as antipyrin may give a leucocytosis. The leucocyte increase of pilocarpine is rather a lymphocytosis. Cinnamate of soda, sodium nucleate, bacterin injections and turpentine have been used in kala-azar to increase leucocytes.

LYMPHOCYTOSIS

Of course, the disease in which we have the most marked lymphocytosis is lymphatic leucaemia.

The lymphocytosis of typhoid fever has been taken up under leucopenia.

Whooping-cough may give a lymphocytosis of 20,000 to 30,000.

Young children have normally an excessive proportion of lymphocytes even to a reversal of the polymorphonuclear-lymphocyte relation of adults. This is apt to be particularly marked in hereditary syphilis. Enlarged tonsils may give rise to lymphocytosis of 10,000 to 15,000 when more than 50% of the white cells will be lymphocytes. Rickets and scurvy give a lymphocytosis.

In pellagra there is a moderate lymphocytosis, averaging 34% in about a normal count.

Varicella and mumps may also give an increase in the percentage of lymphocytes.

Malta fever is a disease which may show quite a lymphocyte increase, this going with a reduction in polymorphonuclears.

_Glandular fever_ (Pfeiffer, 1889) is a mild acute febrile disease, the fever coming on after a short incubation period and lasting about one week. Its main characteristics,—soft enlargement of the lymphatic glands, splenomegaly, and a leucocytosis of about 20,000 with 80% lymphocytes of the lymphoblastic type and many with bilobed Rieder nuclei,—lead often to its being mistaken for lymphatic leucaemia. Throat infections, particularly Vincent’s spirillosis, are thought by some to be concerned in its genesis.

INCREASED LARGE MONONUCLEARS

In tropical work we combine the large mononuclears and transitionals in a differential count. They are the phagocytes of animal cells or parasites. The disease in which their increase is best recognized is malaria and an increase to 15% where the blood shows moderate leucopenia is very significant. The melaniferous leucocytes of malaria are cells of this type.

Other protozoal infections, as kala-azar, trypanosomiasis and amoebiasis cause it. Filterable-virus diseases may show a mononuclear increase, thus yellow fever and dengue both give an increase about the fifth or sixth day.

In Banti’s disease there is an increase in cells of this type and a transitional increase is reported for Hodgkin’s disease.

DISEASES IN WHICH THERE IS A NORMAL LEUCOCYTE COUNT

Uncomplicated tuberculosis, influenza, Malta fever, measles, trypanosomiasis, malaria, syphilis, and chlorosis.

In malaria we have a leucocytosis at the time of the rigor, while during the apyrexial period there is a moderate leucopenia. In malaria we have a marked increase in the percentage of the large mononuclears and transitionals. These may form from 20% to 30% of the leucocytes. When bearing particles of pigment they are known as melaniferous leucocytes—macrophages which have ingested malarial material. In dengue, at the time of the terminal rash, we may have as great a percentage of large mononuclears. In this disease, however, we have a great diminution of polymorphonuclears from the start (25 to 40%). Instead of a large mononuclear we have at the onset a lymphocytic increase. There is an increase of large mononuclears in trypanosomiasis.

The white count is about normal in uncinariasis (Ashford’s average was 7800). Some have reported a leucopenia in severe cases.

While eosinophilia is the most marked feature in hookworm disease yet in very severe cases it may be absent.

Coagulation Rate of Blood

This determination is of value in connection with operations on jaundiced patients.

Wright’s coagulometer is a standard instrument but is cumbersome.

A simple method of determining the rate is to take a piece of capillary glass tubing and hold it downward from the puncture to let it fill for 3 or 4 inches. Then at intervals of thirty seconds scratch with a file the capillary tubing at short distances and break off between the fingers. When coagulation has taken place a long worm-like coagulum is obtained. Normally coagulation occurs in about three to four minutes, when the temperature is that of the hand in which the tubes are conveniently held. Rudolf recommends placing the tubes in metal tube-containers in a Thermos bottle at 20°C. He gives the normal coagulation rate for this temperature as eight minutes, while at a temperature below this the period is lengthened. Age and sex do not influence the rate. Sabrazes, the originator of this method found no appreciable variation in tubes from 0.8 to 1.2 mm diameter.

In Burker’s test you mix a drop of blood in a drop of distilled water on a slide and with a capillary tube sealed off at the end stir the mixture every half minute. So soon as fibrin threads appear you have coagulation.

For the proper testing for coagulation rate the blood should be taken from vein and not from that exuding from a needle stab of ear or finger. Our experience shows that it is not necessary to use venous blood.

Specific Gravity of the Blood

Hammerschlag has a method for the determination of the Hb. percentage based upon the specific gravity of the blood.

In this method a mixture of benzol and chloroform is made of a specific gravity of about 1050. A medium size drop of blood is then taken up with a pipette and dropped into the mixture. If it sinks add more chloroform from a dropping bottle, if it tends to rise, more benzol. The mixture in which the drop of blood tends to remain stationary, near the top of the mixed benzol and chloroform, has the same specific gravity as that of the blood. This is determined by an accurately graduated hydrometer. The normal average specific gravity for men is 1059, for women 1056. A table, giving the Hb. percentage corresponding to the specific gravity, accompanies the outfit.

To determine the necessity for intravenous infusion in cholera Rogers has recently recommended the employment of small bottles containing aqueous solution of glycerine with specific gravities varying from 1048 to 1070, increasing the specific gravity in each successive bottle by 2°.

An accurate hydrometer will suffice to determine the specific gravity. Drops of blood from the cholera patient are deposited at the center of the surface of the fluid in the bottles from a capillary pipette. If the specific gravity of the blood is 1062 at least a liter of saline or sodium bicarbonate solution is needed. If 1066 at least two liters. Formerly he estimated the indications by blood pressure considering a pressure of 80 in Europeans or of 70 in natives as indicating intravenous injections.

PRACTICAL APPLICATION OF METHODS OF BLOOD EXAMINATIONS TO THE VARIOUS TROPICAL DISEASES

In considering the value of blood examinations in the various tropical diseases we may _first_ note those in which such examinations are of little or no value and _second_ those in which such examinations are crucial or at any rate of prime importance.

1. IN THE FIRST GROUP WE MAY INCLUDE THE FOLLOWING:

_Beriberi._—The leucocytes are about normal in number with possibly a slight increase in lymphocytes. Of course there may be anaemia present with the progress of the disease. Some think there is a slight diminution from the normal percentage of eosinophiles.

Noc found the percentage of lymphocytes in beriberi patients to be about 35 as against 32 for those unaffected.

_Sprue._—There is considerable reduction in red cells which may fall below 2,000,000 in advanced cases. The whites may show a slight tendency to leucopenia with a relative increase in lymphocytes. The haemoglobin is not as much reduced as the red cells so that we obtain a color index of from 1.1 to 1.3.

Poikilocytosis and punctate basophilia are often noted, but rarely does one find nucleated reds. In a severe case the blood picture is rather that of an aplastic anaemia than a typical pernicious anaemia. The eosinophiles are rare or absent as the case advances. One often finds many (7-9) nodes in the polymorphonuclears.

_Pellagra._—This disease may show a chloranaemia. Some authorities have stated that we have an increase in the percentage of large mononuclears but Hillman found a rather definite increase in the lymphocytes (34%) and a normal large mononuclear percentage.

_Yaws._—This disease may show a moderate anaemia with a low color index. The leucocytes are about normal in number with a moderate increase in the percentage of large mononuclears.

_Leprosy._—There is, as would be expected, with the progress of the disease, an anaemia which is of the chlorotic type. Leprosy bacilli may be found in the blood, especially during the time of the febrile accessions, but such examinations are of very little value in practical diagnosis and there are so many liabilities to error, as shown in the work with tubercle bacilli in blood, that we should be very conservative in this direction.

There is probably an increase in the percentage of lymphocytes.

_Yellow Fever._—The blood findings are usually given as normal although Noc states that at first we have an increase in polymorphonuclear percentage to be followed by an increase in the large mononuclears about the fifth day. He also noted an absence or diminution of eosinophiles.

Intraperitoneal inoculation of animals with blood from patient should be practised. Should the diagnostic reliability of the procedure be established, yellow fever should then be placed in Group 2, among those diseases in which examinations of the blood are of prime importance.

_Cholera._—As cyanosis develops the red count goes up even to 8,000,000 with a corresponding or greater increase in the leucocyte count. The estimation of the low blood pressure is important as indicating the necessity for intravenous injections. The determination of the degree of serum acidosis is also indicated with reference to alkaline treatment. In a convalescent from a disease suspected as cholera an agglutination test would be of value, and in the absence of the serum of immunized animals one could use that of a cholera convalescent against a spirillum isolated from the stool of a suspected case of cholera.

2. OF THE DISEASES IN WHICH AN EXAMINATION OF THE BLOOD SHOULD ALWAYS PLAY A PART IN DIAGNOSIS MAY BE NOTED THE FOLLOWING:

_Malaria._—The examination of the blood is necessary not only to prove the existence of a malarial infection but, as well, to determine the species of parasite present, this latter a matter of much importance as to prognosis and intensity of treatment according as one has to deal with a benign or malignant parasite. More exact information (and with the expenditure of much less time) can be obtained from a smear stained with some Romanowsky modification than by examining a fresh preparation.

At the same time it is advisable to make a wet preparation and study it for amoeboid activity of the parasites and character of the pigment while awaiting the completion of the staining process.

In the blood of a malarial anaemia the central vacuolation of many of the red cells may give an appearance of young nonpigmented parasites. Malarial parasites tend to move about to take peripheral locations and furthermore they do not change in size upon focussing up and down as do the vacuoles.

Melaniferous leucocytes can be made out better in a fresh specimen than in a dried, stained one.

One can better differentiate species by an even thin film than by a thick-film method. There is often great doubt with a thick film as to whether the object noted is an artifact or a parasite. The Ruge thick-film method has given very good results.

There is only a moderate variation from a normal white count but in cases when the parasites are very scanty or when they have been driven from the peripheral circulation by quinine treatment we may make a tentative diagnosis of malaria on a leucocytosis during the paroxysm with a leucopenia during the afebrile interval with, at this time, an increase in the percentage of large mononuclears to 10 to 15%.

Melaniferous leucocytes are rarely noted in the benign tertian infections but in some of the very puzzling aestivo-autumnal fevers they may give the diagnostic clue.

Schüffner’s dots are yellowish dots in the infected red cells and are characteristic of benign tertian. The Maurer clefts of malignant tertian are less commonly noted. Always carefully note the pale, swollen, infected red cells of benign tertian, the shrunken degenerated cell of malignant tertian and the normal one of quartan. The fine hair-like ring of malignant tertian is often noted on the periphery of the red cell as a narrow line while the half-grown schizont of quartan is often seen as an equatorial band.

In the anaemia following malaria we may have very low red counts and haemoglobin percentages. They usually run parallel, so that the color index approximates 1.

Punctate basophilia is quite common in malarial anaemias. Up to the present time the culturing of the parasite can scarcely be considered an aid to diagnosis as it is difficult to carry the development beyond one generation so that we do not get multiplication of parasites. In cases where confusion exists as to the nature of the species of parasite present culturing would help as regards the possibility of noting the developmental stages of _Plasmodium falciparum_.

_Blackwater Fever._—The same points which hold for malaria hold for blackwater fever.

The striking feature of blackwater, from the side of the blood, is the rapid and great reduction in red cells and haemoglobin. As a result of the pathognomonic haemoglobinuria we may have in a few days a fall of red cells from 4 or 5 million to approximately 1 million with haemoglobin down to 20%. The color index is usually about 1. The blood is thin and the serum tinged. Probably from the excessive haemolysis one does not see degenerated cells as frequently as would be expected. Tests for acidosis and coagulability of the blood are indicated as there is a reduction in titrable alkalinity of the serum and coagulation rate.

_Oroya Fever._—This disease, within two or three weeks, gives the blood picture of a marked pernicious anaemia. The rod-shaped protozoon may be seen lying in the red cells singly or in V-shapes.

These rods show a chromatin granule at one extremity. Normoblasts are very numerous and megaloblasts appear later. There is both polychromatophilia and poikilocytosis. The color index is that of pernicious anaemia, above 1. The leucocytes are increased to about 20,000 with 75% of neutrophiles, among which are many immature forms or metamyelocytes. The pathological process shows its greatest activity in the bone marrow.

_Malta Fever._—In this disease blood cultures offer the surest and most practical way of making the diagnosis. The blood should be taken from a vein at the time of the height of the fever rise. To prevent coagulation the blood should be forced from the syringe into about an equal amount of citrated salt solution and subsequently added to melted agar to then be poured into Petri dishes. Cultures can also be made by smearing the citrated blood over poured plates of agar.

It must be remembered that the colonies are quite small and do not develop for four or five days.

The citrated blood can also be added to bouillon. The blood culturing has rather replaced the culturing from spleen juice. As the coccus is in the blood it is eliminated in the urine and plates should be made from the urine as well as the blood.

Malta fever is one of the diseases which can be diagnosed quite early by agglutination tests, the reaction often appearing before the end of the first week and often continuing for months after recovery. There is a liability to error when low dilutions are employed so that the former use of dilutions of 1 to 20 and 1 to 40 is no longer advised. Probably a dilution of 1 to 100 would be sufficiently specific but dilutions of 1 to 500 and even higher are frequently obtained. It is now thought best to heat the patient’s serum to 56°C. for twenty minutes before applying the test so as to destroy nonspecific agglutinins. Opsonic index and complement fixation tests have been employed in diagnosis.

As the disease progresses a secondary anaemia develops. The white count is about normal but with the polymorphonuclears somewhat reduced in percentage and the mononuclears increased.

Some observers have reported a leucopenia as of some diagnostic value but others find the leucocyte count normal and Rogers considers the absence of leucopenia as differentiating kala-azar from Malta fever.

_Plague._—In septicaemic plague blood cultures offer the surest method of diagnosis as clinically there may be very little to suggest plague. This is about the only disease in which one may find the causative bacterium in a blood smear. For this examination the thick-film method has been recommended. Just as with the material from a puncture of a bubo or the sputum from plague pneumonia we should employ animal inoculation as well as cultural procedures with the blood.

We usually have a marked leucocytosis due to a great increase in the polymorphonuclears. The white count may exceed 50,000. Just as septicaemic plague may so overwhelm the organism that it does not respond with fever so may the leucocytosis be absent. Bubonic and pneumonic plague tend to become septicaemic, so that in such types of the disease we may obtain results with blood cultures.

_Liver Abscess._—Schilling-Torgau brings out the point that even with an absence of the usual blood findings it is possible to diagnose the disease and make a just prognosis with his method of differential counting. Ordinarily we have a leucocytosis of from twelve to twenty thousand with only about 70% of polymorphonuclears and about 12 to 15% of large mononuclears. When a bacterial infection accompanies the amoebic one of course the leucocytosis and polymorphonuclear percentage reach higher figures. The eosinophiles may entirely disappear in an uncomplicated case of amoebic abscess.

In comparing his method with the ordinary one Schilling-Torgau notes a case with a differential count showing 72% of polymorphonuclears, 17% of lymphocytes and 8% of large mononuclears with a white count of 6000—apparently a normal blood. By his method 33% of these neutrophiles were found to be of the band-form or less mature cells, thus showing that the blood really did deviate from the normal.

In other examinations he noted very unfavorable indications from the high percentage of metamyelocytes and even myelocytes when the ordinary count did not suggest the serious condition.

As stated previously this method would seem to offer many advantages over the ordinary one.

_Trypanosomiasis._—While the blood, when examined in ordinary smears or with thick-film methods, does not give as good results as by examining the gland juice for trypanosomes, yet, by taking 5 or 10 cc. of blood in citrated salt solution with 2 or 3 centrifugalizations, we may obtain greater success in finding the parasites in this way than when using gland juice.

In wet preparations we may note the clumping of the red cells. This is the phenomenon of auto-agglutination thought by some to be rather characteristic of trypanosomiasis.

We may carry out the leucocyte attachment test using the inactivated serum of the suspected patient.

As the disease progresses we get a secondary anaemia. The leucocyte count is usually normal but the differential count shows an increase in the large mononuclears. Bacterial infections often supervene when a leucocytosis will be noted.

_Kala-azar._—Quite recently there has been success in the diagnosis of kala-azar by culturing the blood of the suspect on N. N. N. medium. The key to success when culturing from the blood is to wait for two or three weeks before giving up the examination of the cultures. It will be remembered that almost invariably leishman bodies are present in the blood only in extremely small numbers so that there is not time by the end of a few days for sufficient development to have taken place. In probably 80% of cases the parasite of kala-azar may be found in stained smears from the peripheral blood but only after prolonged and patient search. They may be found phagocytized by large mononuclears or polymorphonuclears. Of course splenic puncture examinations show far greater abundance of parasites than blood smears but it is not without danger.

The marked anaemia of kala-azar does not appear until the earlier symptoms of fever and splenic enlargement have gone on for some time. Very characteristic and important in diagnosis, however, is the marked leucopenia of kala-azar, approximating 2000 leucocytes on the average. Again the white cells are only about in the proportion of 1 to 1000 red cells. There is an increase in the percentage of large mononuclears. Some authorities have reported an acidosis of the blood serum. Coagulation rate is delayed.

In kala-azar the coagulability of the serum is altered as shown by the formol-gel test. In this test, a drop of clear serum from the patient is placed on a slide which is then inverted over a watch glass containing a few drops of liquor formaldehyde. In cases of kala-azar the serum will solidify, appearing as an opaque, stiff jelly which adheres to the slide; while other sera will remain fluid, running off the slide when it is tilted. The reaction appears not to be specific since it has been reported for syphilis and other diseases.

_Relapsing Fever._—The spirochaetes are not so numerous in the blood of the peripheral circulation in tropical relapsing fevers as in those of Europe.

The spirochaetes can best be seen in stained smears but the agitation of the red cells in a wet preparation by the motile spiral organisms is of assistance in their recognition. Dark-field illumination, India ink smears and Fontana’s silver method are used as well as Giemsa staining.

During the afebrile period the parasites disappear from the peripheral circulation.

If the disease is first seen during the afebrile stage we may try Lowenthal’s reaction, which consists in taking a drop of the blood of the suspected patient, mixing it on a vaseline ringed slide with the blood of a patient showing spirochaetes, then covering with a cover-glass and incubating for thirty minutes at 37°C. A positive reaction shows clumping and loss of motility of the spirochaetes.

Reports vary as to the white count but on the whole there would seem to be more evidence in favor of a moderate leucocytosis although some observers have noted a fall from the normal. The usual statements give a leucocytosis of from 12 to 15 thousand with a polymorphonuclear increase to between 75 and 80%. The statement is usually made that the normal percentage of large mononuclears helps in the differentiation of malaria. Kieseritzky has reported leucopenia and slight increase in lymphocytes.

The leucocyte count tends to be higher about the time of crisis.

_Weil’s Disease._—This spirochaete infection is due to _Leptospira icterohaemorrhagiae_. The spirochaete has been found in the blood and has possibly been cultured anaerobically from the blood. The practical method is by inoculating guinea pigs with blood or urine sediment. Spirochaetes are found in the liver smears of the sick guinea pigs. In the first week of Weil’s disease we have a leucocytosis—later on a leucopenia.

_Filariasis._—The sheathed embryos of _Filaria bancrofti_ are found in the peripheral circulation at night only, hence _F. nocturna_, while those of _F. loa_ are only to be found in the daytime, hence _F. diurna_. In the islands of the South Pacific the filarial infection is considered as of _F. bancrofti_ but the embryos are present in the peripheral circulation both by day and by night.

Instead of being uncommon it seems rather to be the rule to fail to find embryos in the blood preparations in cases showing marked evidences of filarial disease, as in elephantiasis, calabar swellings, etc. The positive blood findings are most frequent in those who do not as yet show symptoms. There has not yet been sufficient obstruction in the lymphatics to keep the embryos from reaching the blood stream.

In some countries where a large percentage of the population may show embryos in the peripheral circulation, manifestations of the disease are very rare.

We may examine the blood either with fresh preparations, when the movements of the embryos assist in their detection, or by staining dried smears. Haematoxylin staining is better than the Romanowsky one as the break in cells and other points are better brought out.

An eosinophilia is usually considered as constantly present but this is not invariable. The leucocyte count is about normal.

_Dengue and Phlebotomus Fever._—In these diseases a leucopenia, which begins to show itself by the second day, is very characteristic.

The average leucocyte count is about 3500 and along with this we have a reduction in the percentage of polymorphonuclears to about 50%. Towards the end of the terminal fever we have an increase in the percentage of large mononuclears.

_Bacillary Dysentery._—The agglutination tests are of little value in diagnosing the presence of or type of an infection with dysentery bacilli, as the agglutinating power does not appear until during convalescence.

It is now customary to use a polyvalent antidysenteric serum in treatment so that it is not very essential to ascertain the strain involved in an infection. As a practical matter we make our diagnosis of the presence as well as type of dysentery bacillus involved in an infection by isolating the organism from the dysenteric stool.

During the fever we may have a moderate polymorphonuclear leucocytosis.

_Enteric Group of Fevers._—In fevers of atypical course in the tropics one must always remember that _typhoid_ and the _paratyphoid fevers_ are anything but uncommon and blood cultures should always be made when such suspicion arises. In some tropical regions paratyphoid A infections seem most common although the usual experience is to encounter the paratyphoid B infection more frequently. In temperate climates the noting of a moderate leucopenia with an absence of eosinophiles is important in the diagnosis of typhoid, but in the tropics there are so many intestinal parasites and skin infections productive of eosinophilia that we cannot attach any importance to such a finding.

_Typhus Fever._—Plotz attaches importance to the culturing of _B. typhi exanthematici_ from the blood of typhus cases, but the relationship is now regarded as not causal. _Rickettsia_ bodies, which can be demonstrated in the louse or in capillaries at autopsy, are now considered to be the exciting organism.

A mononuclear leucocytosis has at times been reported.

_Spotted Fever of the Rocky Mountains._—Injection of the blood of the patient into guinea pigs produces the disease in the animal. Frick has reported the finding of chromatin-staining bodies in the red cells of such pigs and Wolbach has found chromatin-staining bacteria in the endothelial cells of such animals. These bodies are now classed as _Rickettsia_.

These findings cannot as yet be considered of diagnostic value.

_Various Helminthological Infections._—In the earlier stages of ancylostomiasis and schistosomiasis we have a rather notable increase in the percentage of eosinophiles but with the advanced stages of these infections, with severe anaemia, the eosinophiles may even be absent.

One should always keep in mind the very characteristic and marked eosinophilia of _trichinosis_ when such a blood finding is encountered. There is often a leucocytosis of 15,000 to 20,000 in this disease.

In the _urticarial fever_ stage of _Japanese schistosomiasis_ the marked eosinophilia is of great assistance in diagnosis. One trouble about attaching importance to eosinophilia in the tropics is the confusion which is difficult to eliminate and which arises from infections with the more common but less important group of intestinal parasites such as _Ascaris_, _Trichuris_, etc.

The eosinophilia-producing characteristics of many skin diseases must also be kept in mind.

##