Chapter 19. Laboratory

Clinical laboratory tests have relatively little utility in most resource-poor settings, although lives can be saved if a few simple tests are available, such as those for urine pregnancy, and in endemic/high-prevalence areas, HIV, malaria, and TB screening. This chapter discusses how to improvise various materials, equipment and tests when clinicians need lab tests and the normal procedure for doing them is unavailable.

Table 19-1 lists essential laboratory tests worldwide, which differ depending on the facility's capabilities (described in Table 5-1 in Chapter 5, Basic Equipment).

Water

Use filtered rainwater rather than distilled water to make laboratory reagents, such as stains. Collect rainwater in places where it won't be contaminated, such as in rooftop tanks.1 Besides laboratory use, distilled water is best for use in autoclaves and in pressure cookers used as sterilizers. If available, it should also be used to produce IV fluids.

Distillation involves boiling water and collecting the steam, which then condenses back to water. The water from condensed vapor does not contain salt and other impurities. In austere situations, improvising distillation systems saves many lives, as is seen in the following story from a World War II prison camp:2

Some six weeks before the outbreak I had constructed, as a precaution, a small plant for distilling water for intravenous saline. It was a very primitive affair, the condenser being a coiled rubber tube inside a hollow bamboo in which cold water was circulated. The boiler was a 4-gallon kerosene can. Within 24 hours of the outbreak we were able to give intravenous saline. The plant was producing 40 pints daily, and eventually double that figure. The giving IV bottle was a jam-jar connected to a needle by a rubber tube.

If distilled water is required, it can be easily obtained using a tea kettle and a collecting jar (Fig. 19-1). Heat water in the kettle and run rubber tubing from the spout through a hole in the lid of a collecting jar. Use another small tube as a vent in the jar's lid. Seal all joints with adhesive tape, clay, or glue. Immerse the collecting jar in a pan of cool water so the steam will condense quickly. Voilà, distilled water!

An alternative method of distilling small amounts of water is to fill a large pot one-half to three-quarters full of water. Invert the lid and suspend a heat-resistant cup from a cradle made from wires (coat hangers may be used) attached to the lid's handle. The cup should hang right-side up without dangling in the water when the lid is upside-down. This takes about 5 minutes to set up. As the water boils and steam collects on the lid, it drips down the handle into the cup.4 If only a little water is used, watch the pot more closely (Fig. 19-2).

Power

In austere situations the electrical supply is often intermittent, if available at all. Suitable AC current for instruments can be obtained from rechargeable batteries using a solar or DC/AC inverter.

Microscope

Having a microscope greatly increases one's ability to do laboratory tests. In a World War II prison camp, a Dutch physician, Dr. Bras, "arrived in camp with a crude microscope constructed of bamboo tubing and field-glass lens."5 That takes a lot of skill and ingenuity.

When the electricity fails, use a mirror and sunlight or a battery-operated light as your light source.

Filters

If you need to filter solutions, including stains, use a flower pot with the hole in the bottom. Loosely plug the hole with cotton, fill the pot with several inches of sand, and pour in the liquid. Collect the filtered solution in a jar placed under the pot (put it up on a stand). Replace the sand frequently.

Fire-Fighting Equipment

To quash laboratory fires, keep buckets of sand available to put out small fires. Water may not be readily available (or appropriate for all laboratory fires) and fire extinguishers are both expensive and need regular maintenance.

Lab Burner

Speaking of fires, a simple laboratory burner can be made from a shoe polish container. Make a hole in the lid and fill the container with denatured or wood alcohol. If desired, a small piece of tin can be soldered to the lid to hold a cloth or cotton wick.

Refrigeration

Refrigeration may be needed for some samples and reagents. Improvised refrigeration methods and equipment are discussed in Chapter 5, Basic Equipment.

Flasks

Laboratory flasks can be made from dead lightbulbs. Spotlight-type bulbs are the best, since they have flat "bottoms." If regular lightbulbs are used, an egg carton can serve as a holder. When doing this, wear gloves and eye protection, and cover the bulb's glass with a cloth.

With either type of bulb, first use a pair of needle-nose pliers to pull off the little black button that holds the bulb's electrical contact (Fig. 19-3, left). Doing so leaves a hole in the bulb's porcelain base. Put one of the pliers' jaws into that hole and gently work it around until the porcelain cracks (Fig. 19-3, center). Remove the pieces. The hole (with the filament) becomes visible.

After gently breaking off any protruding glass pieces in the hole, sharply tap the glass piece holding the filament (Fig. 19-3, right) to break it, and it will fall into the bulb. Turn the bulb over and dump the pieces out. A thin metal disc may remain in the bulb. Grab it with the pliers, bend it in two, and remove it. Then bend down any sharp edges on the metal (screw part) that remains on the bulb, rinse it out, and begin using it (Fig. 19-4).

This whole process takes about 5 minutes. Use a known amount of liquid to measure the flask's volume and mark it on the outside at convenient increments (e.g., 50 mL, 100 mL).

Test Tube Holder

To hold test tubes over a burner, fashion a test tube holder by bending a wire hanger or similar material (Fig. 19-5). Use pliers (rather than just your fingers) to do the bending.

Scales

Tiny Scale

To measure small quantities or objects, build a scale using a jar lid and a heavy rubber band (Fig. 19-6). Insert a nail in an upright piece of wood. Make four holes that are equidistant around the lid, attach a wire to each, and tie or twist all the ends together. (Sutures or fishing line may also be used in place of wire.) Use a paper clip (unwound to an "S" shape) to join the wires and the rubber band. Now that the scale has been assembled, it must be calibrated. Attach a piece of tape that you can write on to the upright wood, and mark the gradations as you add small known weights. (Small quantities of water can also be used.) Once the scale is calibrated, either keep the original set of weights or find small stones (mark each stone with its weight) that can be used to recalibrate this scale. Change the rubber band and recalibrate the scale daily.

Heavy-Duty Scale

Used for heavier weights, this scale holds larger items and measures in kilograms. Attach a heavy spring from a chair or automobile cushion to a solid wooden base. Affix two upright posts on opposite sides (Fig. 19-7). These steady the structure and can be marked with incremental measurements when the scale is calibrated. Affix a metal pan to the top of the spring, either by soldering it to the spring or by passing wires through the pan onto the spring. Calibrate the scale by putting objects of known weights on it and marking the slat as the spring is depressed. Known quantities of water (1 kg = 1 L = 2.2 lb) work well. Use the top of the pan as the marking point.

"Steelyard" Scale

Several types of "steelyard" scales can easily be made. Make the long beam from either wood or metal; use pieces of metal pipe for the sliding counterweights. Fasten the beam and the pan together with wire. For support, place the entire balance on a stiff piece of wire that goes either through or around the beam (Fig. 19-8). As with the other scales, use known weights to calibrate this scale.

Incubator and Culture Media

Making media for bacteriology is a complex process; so, if you really need to grow cultures, buy the prepackaged powder.

To build an incubator, get a 20-gallon glass aquarium or any container having the same volume (Fig. 19-9). A cardboard box will do in a pinch, but it may not last long. Turn the container on its side with the open end facing front. Tape a piece of heavy plastic sheeting to the top, draping it down so that it covers the open end. The sheeting is the incubator's "door." A bacteriology incubator is normally kept at body temperature, 98.6°F (37°C). Tape a standard thermometer to the inside of the container so that you can easily read it. Set a standard incandescent lamp enclosed in a can or small pail inside the incubator with the electrical cord going out the "door" to a plug. Begin with a 40-watt bulb and assess the temperature after 12 to 24 hours. If it needs to be warmer, use a higher-wattage bulb. It will probably not be exactly 98.6°F, but if it is a little cooler, that will still work. To be more accurate, use a higher-watt bulb with a dimmer switch to modulate the emitted heat.9

Centrifuge

Centrifuges separate liquid suspensions into their component parts, such as consolidating urine sediment or separating blood cells from serum. They are useful for laboratory work.

If possible, procure a real centrifuge. The difficultly with building a centrifuge is getting it balanced correctly. If it is not balanced, not only does it make a lot of noise, but it will also self-destruct. And although improvised machines often will not produce the rotational speeds needed for some tests, even the simplest centrifuge can cause significant damage if the tubes break or fly out of the machine while it is spinning. Given those caveats, relatively simple and workable centrifuges can be made if necessary.

Hand-Drill Centrifuge

A hand-powered centrifuge can be built using a wooden spoon, a hand drill, plastic test tubes, small eye-screws (with 0.5-inch eyes), and cotter pins (Fig. 19-10). It is very rudimentary, but can be used to spin urine, small aliquots of blood, and similar specimens.10

1. Cut two 1-inch pieces off the end of the spoon handle. These are the "stoppers."

2. Check to be certain that the spoon's handle fits securely into the drill—where a bit would go. If it is too big, shave it down until it fits.

3. At this point, if you have an extra piece of wood with a wide base, drill a hole into it with a little larger diameter than the spoon handle. This can serve as a stand while completing the rest of the construction, and also later when working with fluid-filled tubes.

4. Link the two pairs of eye-screws by opening one of each pair, inserting the other, and closing it again.

5. Drill tiny guide holes in the top of the stoppers and twist an eye-screw from each pair into that hole.

6. Drill little holes through the plastic test tubes, 0.5 inch from the top. They must be exactly opposite to each other at the widest part of the tube (180 degrees apart).

7. Insert the stoppers into the test tubes and mark where the test tube holes meet the stopper.

8. Drill a small hole very carefully through each stopper from the mark on one side to the mark on the other side.

9. Put the stoppers into the test tubes and insert the locking pins. If they fit, remove them for the moment.

10. Screw the other eye-screw from each pair into opposite sides of the widest part of the spoon.

11. Reinsert the stoppers into the tubes, insert the cotter pins through the holes, and bend one side of each cotter pin so that it forms a locking pin for the test tubes.

12. If a vice is available, use it to secure the drill handle. This makes the centrifuge much easier to use. If no vice is available, steady the drill base on a hard surface before operating your centrifuge.

13. Gently rotate the drill handle.

14. Caution: Significant centrifugal force is created. Wear goggles, and if the centrifuge can be operated inside a container (put it in a large cardboard box and make holes to stick your hands through), it would be best. Note that this is dangerous enough with a hand drill. Do not use a power drill.

15. If the system works, add water to the test tubes and try it again.

16. Still okay? Then add your test materials.

17. If you have slightly larger amounts of material, use two identical small plastic screw-top jars in place of the test tubes. Drill holes in their lids the same size as the test tube stoppers. When using jars, put the stoppers through these holes; the cotter pins go underneath the lids.

Table 19-2 contains a variety of conversion factors needed when working in different locales.

Diagnoses Without a Lab

If laboratory tests (or imaging) are not available, the evaluation of sputum volume, stratification, color, consistency, and odor can help make the diagnosis.

Volume

No more than 20 mL of sputum is normally produced daily. An increase in volume to ≥100 mL/24 hours occurs in (a) pulmonary edema, (b) bronchiectasis, (c) chronic bronchitis, (d) advanced pulmonary tuberculosis (TB), (e) lung abscess, (f) asthma, (g) pulmonary gangrene, and (h) extrapulmonary lesions (e.g., amoebic liver abscess bursting into the lung). Observing daily sputum volume is a good indicator of deterioration or improvement of lung conditions. Sudden cessation of sputum is highly suggestive of bronchial plugging.11

Stratification

Sputum from many patients with lung infections separates into three layers when left standing in a tall narrow container; the top layer is frothy, the middle is faintly turbid, and the bottom layer consists of tissue debris. Such layering occurs in (a) bronchiectasis, (b) lung abscess, and (c) gangrene of the lung.11

Color and Consistency

Various disease processes and organisms will produce characteristic changes in the color and consistency of sputum.11 This can assist clinicians in making a diagnosis.

Odor

A putrid odor is associated with bronchiectasis and lung abscess.11

Tuberculosis Testing

The laboratory exam for TB should begin with an acid-fast stain (read by someone who is not color blind). If that is negative, concentrate the sputum and stain the concentrate using the acid-fast method. Without sophisticated laboratories, that is the most that can be done.

Concentrating Sputum

Using concentrated sputum to test for TB greatly improves the sensitivity of staining for the organism. In austere situations, a centrifuge may not be available to perform the typical concentration technique used in labs. If an initial stain for TB is negative, use the overnight sedimentation technique discussed below. This procedure increases the sensitivity of direct TB smears as much as standard centrifugation techniques.12

Collect three sputum samples. While these have traditionally been obtained on successive days, if they are all collected at different times on the same day they are just as likely to contain the organism.13 Put the sample in a flask and add twice that volume of a solution containing 3% ammonium sulfate and 1% sodium hydroxide. Shake the mixture by hand and let it stand overnight (≥12 hours) at room temperature. After sedimentation occurs, decant the supernatant fluids and make smears from the sediment. The sensitivity of the smears against the culture is 58% for the direct method and 81% for the concentration method.14

Using a similar method, add 5% sodium hypochlorite (bleach) to an equal quantity of the sputum, agitate by hand for 10 minutes, and then add 4 times the amount of distilled water and let stand overnight. As an example, add 1 mL 5% sodium hypochlorite to 1 mL sputum. After 10 minutes, add 8 mL distilled water and let stand for ≥12 hours. This method is statistically equivalent to the classic concentration and centrifugation methods—and much cheaper and easier to do.15

Hemoglobin Measurement

WHO-sponsored, commercially available colorimetric determination kits have been shown to be accurate for estimating hemoglobin (Hgb) levels, particularly in neonates and young infants.16 No special skills are necessary except for being able to obtain a drop of blood. Normal color vision and color interpretation are also important.

The kits, containing 200 to 1000 test strips, include a small card with six shades of red representing Hgb levels of 4, 6, 8, 10, 12, and 14 g/dL. The results indicate whether the patient is anemic, and can be used to decide whether the patient needs a blood transfusion, formal blood count, or referral. Results cannot be used to identify minor changes in Hgb during treatment.

To use the kit, place a drop of blood on the test strip (~2 ¢ each). Wait 30 seconds and immediately compare the strip to the Haemoglobin Colour Scale that accompanies each kit. The comparison must be made in bright light. Training nonprofessionals to use the kit takes about 30 minutes.17 One benefit of this kit is that it also screens out sickle cell anemia, since these patients always have grave anemia; a Hgb level above 8 G% (Hct 24%) excludes the disease.18

Coagulation Tests

While more sophisticated coagulation tests will not be available in resource-poor settings, health care workers can easily do a bedside clotting time, which is vital to determine which patients need snake (Crotalid) antivenin after envenomations. The less-useful bleeding time test can also be easily done.

Clotting Time

Put 1 to 2 mL of blood into a new, never-washed test tube and let it sit for 20 minutes. At 20 minutes, tip the tube upside down once. If no clot is seen, the test is positive (they are not clotting sufficiently). If any clot is present, even a very small one, the test is negative.19 In some remote settings, the syringe in which the blood sample was drawn is used in place of a test tube; in other cases, the blood is injected into an empty ceftriaxone bottle. Limited studies show that these three methods produce congruent results.

Bleeding Time

Doing a bedside bleeding time may have some utility for: (a) determining the cause for ongoing bleeding; (b) explaining previous bleeding episodes; and (c) diagnosing hereditary bleeding disorders, especially von Willebrand's disease.20 However, without standardization, the results must be interpreted cautiously.

To perform the test, seat the patient with his elbow slightly flexed and his forearm resting on a steady support with the volar surface exposed. Place a sphygmomanometer above the antecubital fossa and clean the forearm with alcohol, letting it air dry. Inflate the cuff to 40 mm Hg and wait 30 seconds. Holding a sterile #11 scalpel blade at right angles to the skin, make two 1-mm-deep punctures about 1.5 cm from each other, parallel to the antecubital crease, and about 5 cm distal to the antecubital fossa. Avoid cutting superficial veins. Begin timing immediately after the incisions are made. Blot the incisions with filter paper every 30 seconds until blood no longer stains the filter paper. Do not blot the wound edges. The bleeding time is the average of the time it takes for each of the two cuts to stop bleeding. A normal value is <11 minutes.21

Color

Normal urine varies from yellow to amber, depending on its concentration. Edible dyes, including those in medications and certain illnesses, can change its color (Table 19-3).

Urinalysis

Urine is easily tested with multifunction sticks. These can test for the presence of protein, glucose, ketones, nitrites, bilirubin, urobilinogen, red blood cells (RBCs) or free Hgb, white blood cells (WBCs), leukocyte esterase, pH, and specific gravity, depending on the stick used. The strips must also not have deteriorated due to moisture or temperature.

Cloudy urine can be due to a variety of reasons. Usually, it is due to the presence of blood (generally pink or red), crystals, or infection. If signs and symptoms of infection are present, do the three-glass test to improve the chance of localizing the infection. During a single void, collect the first 5 mL in one container. Collect nearly all the rest of the urine in a second container. In the third container, collect the last 5 mL or so. If the first container is very cloudy, but the urine clears in the second and third containers, the infection is probably urethral. If the second and third containers are cloudier than the first, the infection probably resides in the bladder or kidney. If the third container is the cloudiest, the prostate is probably the culprit.

Rather than spinning urine, with or without a Gram stain, an easier method to determine the presence of infection in a good urine sample is to look for ≥1 bacterium/HPF (high-powered field). That correlates well with >100,000 organisms/mL if the same specimen is cultured.

Taste

Even the ancients knew that urine that tastes like sugar indicates diabetes mellitus. Go ahead—taste it. That's real austere medicine. (I prefer to use the test strips.)

Protein

If a urine dipstick to measure protein is not available, another way to check for proteinuria is to fill a test tube three-quarters full with a clean-catch specimen and heat the top half of the tube over a low flame. Use metal tongs to hold the tube, and rotate the tube while heating it so that the glass does not shatter. If the urine turns cloudy and white, add a few drops of vinegar (2% acetic acid). If it stays cloudy or gets whiter, there is protein in the urine.22 Protein can appear in the urine due to infection, bleeding, intrinsic renal disorders, or preeclampsia.

Gram Stain

Gram stains of urine specimens are done either on spun sediment or on bacteria from urine cultures. Gram-staining slides can provide a tentative identification of bacteria in urine, pus, sputum, cerebral spinal fluid (CSF), and bacterial cultures. Although not highly accurate for identifying species, a Gram stain enables the clinician to make an educated guess about the appropriate antibiotic to use, if the information is combined with knowledge of the clinical situation. The test requires several chemical solutions to prepare the slide and a microscope to view the end product.

The basic "sure-fire" technique is to make a slightly uneven smear from the specimen. Unless you are in a grave hurry, air-dry rather than heat-fix the slide. Wear gloves to avoid staining your hands rather than using forceps to handle—and often drop—the slide. Once the slide is dry, cover the slide with ammonium oxalate crystal violet and wash off the back (not the front) of the slide using a thin stream of water. While smears of pus are relatively durable, urine sediment is fragile and, if washed directly, will go down the drain. Then, do the same with Gram's iodine solution. The amount of time the stains are on the slide (3 seconds to 3 minutes) does not matter—but they must be identical. Then, hold the slide with one end slanting down, and drip on the acetone-alcohol until the color draining from the slide sharply decreases. Immediately wash the back of the slide. Pour on the safranin and immediately wash it off. Dry the back of the slide and, if in a hurry, dry the front by waving the back of the slide over a Bunsen burner, alcohol lamp, or similar heat source. With uneven smears, WBCs in the thin areas will have pink nuclei, but they will be bluish or purple in the thicker areas. Note that the organisms may vary in color. That is usually because, unlike laboratory specimens grown at the same time, "bugs" from real specimens are at different stages in their life cycle, and hence will have different colors.23

Draw conclusions with care—accuracy depends on training and skill. Gram-positive bacteria may appear gram-negative if they are old, if the patient has been treated with antibiotics, or if you washed the stain off the slide. Artifacts, such as particles of crystal violet, may be misinterpreted as cocci or bacilli. Some material from the smear may have disappeared during staining if you did not fix it properly or if you made the smear too thick.

Parasite Exam

The presence of parasites in stools is common, especially in austere situations. Microscopic examination for a wide variety of intestinal parasites is easier and more successful using the following technique than with other common methods.24 Even with this method, however, experts will identify only about 38% of the patients who have parasites. Get other samples if you think the patient has an intestinal parasite.

1. Collect the stool in a container.

2. Mix it with an applicator stick until uniform or homogeneous.

3. Use the same applicator to place a thick circular stool smear over an area 1- to 1.5-cm diameter on a glass slide.

4. If the stool is dry, hard, or well-formed, add a few drops of sterile saline so that it has the consistency of paste.

5. Before the specimen dries, add 1 to 2 drops of lacto-phenol cotton blue (LPCB) stain. This is one of the most commonly available laboratory stains in the world.

6. Examine the slide for ≥5 minutes.

Presence of Mucus

A stool with fresh or altered blood and plenty of mucus is passed in amoebid dysentery. It often has an acidic reaction to litmus paper. Fresh, watery stools with mucus flakes, an alkaline reaction to litmus paper, and an offensive odor accompany bacillary dysentery. "Rice water" and "pea-soup" stools are passed with cholera and typhoid, respectively. Increased mucus or mucus in tape-like form is often passed in mucous colitis. Large frothy stools are associated with pancreatic dysfunction, celiac disease, and tropical sprue.11

A devastating and common disease, bacterial meningitis may be difficult to distinguish from viral meningitis and normal patients—especially with clear or bloody cerebrospinal fluid (CSF). In austere situations, this differentiation is vital, both to conserve resources (antibiotics, hospital beds, personnel time) and to treat those patients who actually have bacterial meningitis.

Bacterial Meningitis Diagnosis

Urine test strips can help distinguish patients with bacterial meningitis from those with viral meningitis or no meningitis, despite the CSF being clear or bloody. That is, they help make the decision to treat patients who might otherwise go untreated even though they have the disease.

Rapid differentiation between bacterial and viral meningitis can be made using three components of typical urine test strips, with a specificity of 100% and sensitivity of 97%. This test helps in making a rapid decision about whether to use antibiotics in patients with suspected meningitis. If adequate CSF is available, the reagent strip is dipped directly into a tube that will not be used for microbiologic tests. Otherwise 1 to 2 drops of CSF are placed on the glucose, protein, and leucocytes portions of the strip. Wash off the strips after 60 seconds and compare the color change against the standard on the test strip container. The presence of leukocytes is graded as negative, 10 to 25, 75, and 500/microL; protein as 30, 100, and 500 g/L; and glucose as <2.8, 2.8, and 5.5 mmol/L.

A diagnosis of viral meningitis is made if the glucose is from 2.8 to 5.5 mmol/L, protein is present, and the leukocyte count is from 10 to 75/microL. Patients with values lower than this do not have meningitis; those with higher values have bacterial meningitis.25 This test, however, will not detect all patients with bacterial meningitis.26 A positive urine nitrite strip test may also help to indicate which patients with turbid, bloody or, most importantly, clear CSF have bacterial meningitis. The test is negative in patients with CSF malaria.27

Protein Evaluation

Without urine test strips, Pandy's test is an alternative method of testing for CSF protein. Prepare Pandy's solution by filling a bottle (~100 mL) one-quarter full with phenol and the other three-quarters with clean water. Shake the bottle and let it stand 24 hours. The two fluids will partially separate, with phenol being at the bottom of the bottle and a phenol-water (Pandy's) solution being at the top. Carefully decant a few drops of the top mixture (avoid including phenol from the bottom layer) into a test tube and add a few drops of the CSF. If the mixture in the tube is cloudy against a dark background, it is positive for excess protein. It becomes positive with 25 to 35 mg/dL protein. While a weak positive may sometimes occur in young children (who may have up to 40 mg/dL protein), a strongly positive test is always abnormal.

CSF Turbidity

CSF turbidity, if present, is one way to identify patients with bacterial meningitis. Although this test may often be falsely negative, checking for CSF turbidity may be the only method of diagnosing meningitis. To have the best chance of identifying turbidity, use clean water to fill an identical tube to the tube with the CSF sample. Compare the two in good light, or at least with a strong light behind them. If the CSF is as clear as the water, there are probably <100 WBCs/microL.22 Note that when looking only for CSF turbidity, one misses a significant number of bacterial meningitis cases.

Home pregnancy tests and home blood glucose tests have become ubiquitous and are generally adequate for clinical use. Also, they are often much less expensive than similar tests sold for professional use. The only caveat is that heat and humidity can make them inaccurate.28 If using them in a health care facility, store them properly. (See Chapter 28, Obstetrics/Gynecology, for additional pregnancy tests.)