Chapter 7. Vital Signs, Measurements, and Triage

Knowing a child's age is important to determine normal vital signs, disease prevalence, and the appropriate medications. It is also an indicator of social milestones. The easiest way to determine a child's age is to ask the parent or check available records. However, in cultures where birth records are unavailable, as well as during acute out-of-hospital or emergency department (ED) events, this information may not be easy to get.

Without records, one way to estimate a child's approximate age is to count the number of teeth present and then add six to derive the age in months.1 Another way is to have the child sit upright (with the head in a neutral position), then raise one arm over his head and try to touch the opposite ear (Fig. 7-1). This "overhead test" has a sensitivity of 90% and a specificity of 78% (positive predictive value 93%; negative predictive value 68%). If the child successfully performs this test, he is most likely ≥6 years old, which is the age to begin school and to receive other benefits in many cultures. If a child fails this test, there is a good chance that he is <6 years old, but it is less clear, since other factors, such as malnutrition, may influence the results. The test works because a 6-year-old child's humerus is long enough to raise the elbow high enough to extend the forearm across the head and allow the fingers to reach the opposite side of the head.2

Any firm surface can be marked in 0.5-cm (or 0.5-inch) increments to measure a patient's height. For ambulatory patients, use a vertical surface, such as a wall or the side of a doorway. For infants, children, and bed-ridden patients, mark the side rails of stretchers and beds with indelible ink. Once a child's height is known, the weight can be estimated.

Using a stretcher marked along the side to measure a child's height is extremely useful; knowing the child's height, you can estimate weight, and this provides the information necessary to calculate pediatric medication and fluid dosages.

If an adult's height can be measured only in a sitting position, such as a wheelchair-dependent patient, measure each arm span in centimeters, from mid-sternum to the end of the extended middle finger (half-arm span). Use the longer of these measurements in the following formula: Height ≅ (0.73 × [2 × half-arm span]) + 43 cm.3 A simpler method is to double the longest measurement from mid-sternum to the tip of the patient's long finger.4

Children

Need to estimate a child's weight? When asked, the parents of children aged from 1 to 11 years old can estimate their child's weight within 10% of the measured weight only 78% of the time. Accuracy when using the Broselow tape was within 10% of the measured weight only 61% of the time, and the accuracy was worse as children got older. At least in developed countries, the Broselow tape tends to underestimate children's weights.5–8 All other forms of guessing a child's weight, including the Argyll, the Advanced Pediatric Life Support, and the Best Guess methods, perform poorly.9 If the parents aren't available, use the following formula:

(2 × age in years) + 8 = weight in kilograms (kg)

Beam Scale for Infants

If standard scales to weigh infants are not available, make a beam scale (Fig. 7-2). If finely calibrated, the beam scale may also be used to weigh smaller items, such as medications.

To make a beam scale, use a straight, rigid pole of bamboo, metal, or dry wood, ~1-m long. Mount a hook close to one end from which the beam will be suspended. Put another hook about 5 cm away, between the first hook and the nearest end of the beam. This is where the infant or substance to be weighed is suspended. A fixed weight is also suspended from the beam, but on a moveable slide (e.g., metal wire). The weight can be a bag, bottle, or other container filled with sand. Mark gradations on the beam for each weight increment. Use standard weights to calibrate the scale initially, so the gradations are accurate. Hang the known weight at the end of the scale (i.e., where the infant would be) and mark the point on the beam where the sliding weight is when the scale balances evenly.10 Large increments (1 L of water weighs 2.2 kg, 0.5 L weighs 1.1 kg, etc.) and smaller increments (use the weight of a common coin and its multiples) can be easily determined. If a local merchant (jeweler, grain dealer, etc.) has standard weights, try to borrow them for more accurate calibration of the scale.

Adults

When experienced health care workers make weight estimates for adults, they are fairly accurate.11 However, the following method for estimating adult weights has been shown to be better than that of experienced physicians, much better than that of nurses, but not as good as the patient's estimate of his own weight. These formulas can be preprogrammed in a computer for easier use.12

Male's weight (kg) = [knee height (cm) × 1.10] + [{MAC (cm) × 3.07} − 75.81]

Female's weight (kg) = [knee height (cm) × 1.101] + [{MAC (cm) × 2.81} − 66.04]

• Knee height: The distance between the thigh prominence with the knee bent at 90 degrees proximally and the plantar aspect of the foot at the heel with the ankle bent at 90 degrees distally.
• MAC (mid-arm circumference): Palpate the humeral head proximally and the lateral humeral condyle distally. Wrap a tape measure around the arm midway between these landmarks so that it contacts the skin at all points but does not compress the soft tissue.

Adults' nutritional status best correlates with body mass index (BMI). The formula to calculate BMI is body weight (kg) divided by [height (m) squared], or [kg ÷ m2].3 For adults, a BMI ≥18.5 is normal; BMI <16 is severe malnutirition.13

In austere circumstances, a thermometer normally is not available. Generally, caregivers notice a tactile fever in only about half of all children with a measurable fever. Elevated tactile temperatures, as measured at home by mothers touching their child's forehead, have moderate (46% to 73%) correlation with fevers recorded in the ED or hospital.14,15 However, for children >2 years old with a temperature ≥38.9°C (102.02°F), tactile assessment can determine that a child has a fever 98% of the time.14

If you do have a thermometer and you need to convert the numbers from Celsius (Centigrade; C) to Fahrenheit (F) or vice versa, use these formulas:

°F = (°C × 1.8) + 32°

°C = (°F − 32°) ÷ 1.8

An Osborn or J wave on an electrocardiogram (ECG) can suggest hypothermia, especially in elderly patients in whom the diagnosis is not suspected. This elevation at the QRS-ST junction (Fig. 7-3) occurs in about 80% of hypothermic patients (a core body temperature ≤95°F[35°C]) and in nearly all patients whose temperature is <90°F (32.2°C). The waves' sizes appear to be inversely related to temperature, but otherwise they do not appear to have any prognostic value.16

Measuring Time

To time pulse or respiration measurements without a watch, clock, or other timepiece, you must either count the pulse beats or respirations yourself or make a timer.

Counting is not as accurate as using a watch, but time may be measured by counting in any situation: Simply count "one-thousand-one, one-thousand-two," and so on, while someone else counts the pulse. To calculate pulse rate per minute, count to "one-thousand-six" (about 6 seconds), then multiply the counted pulsations by 10. Simply estimating time without counting always results in underestimation.

To make a simple, portable sand timer, use a blood collection tube or other piece of glass that can be molded when heated. Heat the middle of the tube and stretch to narrow the diameter. Then, if both ends of the tube are open, seal one end by heating it over a flame. Next, sift some sand through a fine strainer. Dry the sifted sand completely by using sunlight or an oven. Stand the tube upright and put enough sand in the top part so that the sand takes one minute to run from the top, through the narrow middle, to the bottom section (Fig. 7-4). During construction, you will need a watch, a clock, or a previously calibrated timer to check the initial timing. After the correct amount of sand is in the tube, seal the open end either by heating or with a stopper. Since these tubes break easily, store the timer in a padded box.

Water timers also work well. Heat the end of a long, narrow glass tube. Stretch that end and break it, leaving a very tiny opening. A micropipette with a very tiny opening may also suffice without the need to work with heated glass. In either case, make a narrow mark near the top (near the wide opening) of the tube with nail polish, tape, or another marker, or by etching the glass. Cover the narrow end of the tube and fill the tube with water exactly to the level of the mark. Let the water drip out of the tube for 1 minute. Mark the tube again at the exact point of the water level after 1 minute (Fig. 7-5). Voilà, you have a timer!

Finally, if you have no clock or wristwatch for measuring pulse and respirations, use the clock or timer function on the now ubiquitous (even in the least-developed regions of the world) cell phones—if it has a second hand.

Alternate Sites to Palpate Pulse

When you can't palpate a pulse at the typical wrist or femoral areas, try one of the sites listed here:

• Carotid artery. Slide your fingers from the midline to the opposite side of the neck from where you are standing and stop at the anterior border of the sternocleidomastoid muscle. The carotid should be palpable. Don't press too hard and cause a vagal response.
• Superficial temporal artery. This is accessible in nearly all patients. Put two fingers just anterior to the ear at the superior part of the tragus. The finger most proximal to the ear will normally feel the pulsation without difficulty.
• Dorsalis pedis and posterior tibial (foot/ankle) arteries. These are often used to check for peripheral vascular integrity, and also to monitor the pulse, especially when the other sites are inaccessible. These arteries can also be used to draw arterial blood gasses and are much easier and safer sites to draw from than, for example, femoral arteries.

Pulses in Infants

Palpating an infant's pulse, especially when they are critically ill and a decision must be made immediately about whether to begin resuscitation, can be very difficult. The easiest, fastest, and most successful method of determining infant cardiac activity can be applied in any situation, since it requires no equipment. It doesn't even require finding and palpating a pulse: Simply place your ear against the infant's chest wall and listen for heart sounds (Fig. 7-6).

A study with experienced pediatric nurses to find a pulse in infants showed that the mean time to correctly identify the pulse rate using direct auscultation was 2.4 ± 1.2 seconds, with a 100% success rate. Palpation methods took longer and were less successful. The frequently used femoral pulse rate took 9.1 ± 5.9 seconds to find, with only a 43% success rate.17

Radial Pulse and Trauma Prognosis

A weak radial pulse in most trauma patients (at least those 18 to 50 years old without a head injury) indicates a markedly increased mortality (29%) compared to only 3% of those with a normal pulse (odds ratio: 15.2). There is also a strong likelihood that trauma patients will require intubation (odds ratio: 7.5) and even admission to the intensive care unit (ICU) (odds ratio: 5.3). Patients with an absent radial pulse are usually dead or quickly will be dead.18 In addition, tachycardia is a much more reliable sign of hypovolemia than is blood pressure.

"The ability to obtain a blood pressure measurement in an austere environment is often limited by time constraints, equipment availability, and noisy conditions."18

Measuring Blood Pressure Without a Cuff

Estimating blood pressure (BP) when a pulse is palpable, for example, radial (>80 mm Hg), femoral (70 to 80 mm Hg), or carotid (60 to 70 mm Hg), overestimates the patient's actual BP.19

A useful finding is that patients with palpably cool extremities as compared with those of other patients have hypoperfusion. Especially when found in an arm and a leg, such patients have low cardiac indices, low pH, low bicarbonate levels, low mixed venous oxygen saturation (SvO2), and elevated lactate levels.20

Measuring Blood Pressure Without a Stethoscope

When transporting patients by air or in a noisy ambulance, Korotkoff sounds (BP) may not be audible through a stethoscope. If a stethoscope is not readily available or cannot be used, palpate the systolic pressure. If done carefully, this method is as accurate as using a stethoscope.

If a pulse oximeter is available, it also can be used to measure the systolic BP when there is too much noise or vibration either to use a stethoscope or to palpate a pulse. Place the sensor on a finger, get a good waveform, inflate the sphygmomanometer until the waveform disappears, and then slowly deflate the cuff until the waveform reappears. The measurement when it reappears is an accurate measure of the systolic pressure.21

Makeshift/Alternative Blood Pressure Cuff

Neonatal arm BP cuffs can be used successfully on adult forefingers, producing a "finger cuff." Only the mean BP equates to that taken on the arm.22 If a manometer is available, some disposable (or non-disposable) pressure pumps for intravenous (IV) fluids and blood can be used as (a) a BP cuff if attached to a manometer or (b) an extremity tourniquet (clamp the tube when inflated). In large adults, these devices may only fit on their calf or forearm.

Wrist and Calf Blood Pressures

Blood pressure cuffs placed on the forearm (Fig. 7-7) and above the ankle (Fig. 7-8) deliver the same or nearly the same BP as when placed at the brachial artery (arm). Forearm or ankle measurement comes in handy with obese patients on whom the standard adult cuff is too small, for adults when only a child-sized cuff is available, or to avoid compressing an area with trauma. In addition, use a forearm or ankle BP to avoid periodically occluding an arterio-venous fistula (in dialysis patients) or an intravenous line. The mean and diastolic pressure readings are essentially equal in the arm and the ankle, but the systolic pressures are not. So, use the mean BP with a calf BP cuff. (Calculate this by multiplying the diastolic pressure by two and adding the systolic pressure. Then divide the result by three. Alternatively, take one-third of the difference between the systolic and diastolic pressures and add it to the diastolic pressure.)

This difference between the calf and the arm BP readings is reflected in the normal ankle–brachial index (ABI), used to test for vascular compromise in the leg. The calf's systolic pressure is normally higher than that in the arm. (ABI is calculated by dividing the systolic pressure in the arm over that in the leg.) A Doppler is generally used for accuracy. The normal result is <1.23

Preventing an IV Tube from Backing Up

To prevent an IV from backing up and possibly contaminating the IV solution when the BP cuff inflates on the same extremity, run the IV tubing under the cuff, so that when it inflates, the IV stops. When the cuff deflates, the IV starts flowing again. This technique, of course, does not prevent the IV from periodically stopping, as it does when a wrist or an ankle cuff is placed proximal to the IV.24

Children are not small adults; they have the disturbing habit of continually changing their height, weight, and normal vital signs as they grow. So, it helps to have some idea of what these are supposed to be (Tables 7-1a and 7-1b), especially when calculating fluid and medication requirements. This section includes both some simple formulas and more specific tables for various pediatric parameters.

Formulas for Quick Estimates (Normals)

Systolic BP in Children Aged >1 Year (Approximation)25

Median systolic BP = 90 mm Hg + (2 × age in years)

Minimum systolic BP = 70 mm Hg + (2 × age in years)

The Foot as a Vital Sign

In neonates, the sole of the foot may provide a general idea about the child's status. Cold or dusky soles can indicate one or all of the following: (a) hypothermia, (b) hypoxemia, or (c) hypotension, especially in a child with delays in capillary refill and without respiratory distress who does not respond to rewarming.26

Pediatric Milestones in Development

All parents question whether their children are normal. The answer to this question is that no child is "normal," but the child may be developing normally. When assessing the child's development, an easy way of remembering some basic developmental milestones is that a 1-year-old does one action, a 2-year-old does 2 actions, etc. An alternative method is:

The practitioner needs to be able to judge the size of wounds and incisions, as well as skin and internal lesions. Standard measurements, in centimeters or in inches, can be made on personal medical equipment that you normally carry, such as the metal earpieces in a stethoscope (Fig. 7-9), the reflex hammer's handle, penlights, and scissors.

Since antiquity, practitioners have used their own body parts to make these measurements. Table 7-2 gives some approximate lengths and areas for the sites described on one's own body. These correspond most closely to adult Caucasian males. In practice, reasonable approximations such as these usually suffice. However, since body sizes vary greatly, especially between men and women, practitioners should measure these sites on their own body to use for approximations. (Figures 7-10, 7-11, 7-12, 7-13, and 7-14 demonstrate how to take these measurements.) The area measurements are particularly useful for estimating the size of skin lesions.

Volume Measurement

Measuring volume is useful, not only to determine the output of urine and other secretions or bodily losses (such as blood), but also to measure fluids and medications to give patients.

Most food containers (cans, cartons, bottles) have their volume on the label. When empty, they can be used to accurately measure volumes. They generally only need to be cleaned of food or whatever else they previously contained. They need not be sterilized to measure bodily output, unless the specimen also has to go to the lab. Sterility, of course, is vital for anything used to measure material for injection. For items that will be taken orally, they need to be "potable clean," meaning that you could safely drink from them. The volume measures on some containers may not easily equate to those that you know (or need); Table 7-3 provides common conversions.

Kitchen-measuring devices can be used to measure volume. Those with a capacity of ≥32 ounces (oz) (1 L) are especially good for measuring urine output. A transparent container with measurement markings makes the task much easier. The container may be used a lot, so use something durable.28

Most glass or plastic containers normally are not marked with volume increments. Mark them either by fastening a piece of tape from the bottom of the container along its side, or by etching or marking them with an indelible marker. Add 1 mL (or 1 oz) at a time from a syringe or other source of known quantity. Mark the tape or the container at the level reached after each quantity is added. Repeat the process until the container is full.

A patient's overall color can provide valuable information, especially in critical situations or in surgery. Abnormal skin coloring can signify a wide variety of diseases, including jaundice (yellow), plethora (red), and paleness (blanched). One of the most important clues is the presence of cyanosis (blue), often indicating decreased oxygenation. Recognition of central cyanosis (due to deoxygenated hemoglobin) is best done by viewing the tongue. It is more sensitive for this finding than are the earlobes, conjunctivae, or nail beds.29,30 Cyanosis of the tongue also can be recognized in dark-skinned patients, in whom skin color is not useful when making the diagnosis.31

In the operating room (OR),

a patient breathing or being ventilated with air will not look as pink as the "supersaturated" patient familiar to the modern anesthetist using high concentrations of oxygen, but his arterial oxygenation is nonetheless likely to be adequate. Lighting conditions, green drapes, the reflecting quality and color of the walls, vasospasm, and venous congestion due to position can all cause an appearance of cyanosis. When there is any doubt, the color of the open wound is the most reliable clinical indication of oxygenation of the patient.32

The amount of blood a patient has lost determines whether he will receive blood (and how much) for resuscitation. Initially, clinicians may need to estimate this, especially in the ED or other uncontrolled situations. In the operating room, it should be possible to measure at least some of the blood loss.

Estimates

After blood loss, an inability to stand secondary to postural dizziness or a large postural pulse change (≥30 beats/min) strongly suggests that a patient is significantly hypovolemic. After moderate blood loss—or loss over a long period of time (e.g., slow gastrointestinal [GI] bleed)—a patient may not have this symptom. Postural hypotension is much less useful for hypovolemia due to dehydration.33

Excluding closed fractures and stab wounds, the patient's fist and open hand can be used to help estimate blood loss from soft tissue wounds.34,35 The volume of an adult male's fist is about 500 mL (1 unit of whole blood; 2 units of packed erythrocytes); an adult female's fist is between 250 and 350 mL. Use this measurement when the blood lost can be seen or place a fist over areas of the body where there is soft tissue swelling to approximate occult blood loss.36

Likewise, tissue damage equal to one of the patient's fists or skin loss equal to the area of one of the patient's open hands equates to a blood volume loss of ~10% (i.e., 500 mL in an adult, where normal blood volume = 80 mL/kg). This figure includes both the initial loss at the time of injury plus continuing loss through exudation and into tissue (swelling) in the first 48 hours.36 While this is a rough estimate, wounds with more than 5 "hands" of tissue damage lost ≥50% of the total blood volume.34  Table 7-4 gives the approximate blood loss from injuries to various areas of the body.36,37

Measurements

To maintain adequate blood volume during surgery, it is essential to continually assess blood loss throughout the procedure, especially in neonates and children where even a small amount lost can represent a significant proportion of blood volume. Weigh blood-soaked swabs and suction bottles, and estimate the blood lost into surgical drapes and onto the floor; then subtract the volume of irrigation fluids. An alternative is to use the estimate that 10 well-soaked large abdominal swabs represent a blood loss of 1.0 to 1.5 L.37

Repeat Physical Exams

In resource-limited situations, the repeat physical examination remains one of the best tools in the clinician's armamentarium. When the same clinician examines the patient over time, diagnoses for abdominal and, often, chest and neurological conditions become clearer.

Attaching Electrocardiogram Leads

If an electrocardiogram (ECG) or a cardiac monitor is available, but the way of attaching the leads to the patient is missing, several methods work well. The key is to remove any device hiding the bare metal leads (that usually are covered by devices that attach to tape leads on Western ECG machines). After removal, place the leads directly on small saline or alcohol pads in the same limb positions as normal. Affix them using phlebotomy tourniquets. If chest leads are needed, these can be placed on the skin in the same manner, using tape to temporarily secure them. If they must be kept on for some time or if the patient has injuries (e.g., burns) precluding the use of tape, insert small-gauge needles just beneath the epidermis and use alligator clips to make a connection (Fig. 7-15).

Prehospital

A disaster involves multiple casualties that require more resources than are currently available. Triage is the system for quickly allocating the few medical resources.

Patient Classification

Patients are generally classified into one of four groups, depending upon the gravity of their medical problem: Immediate/Emergent (Red), Delayed/Urgent (Yellow), Minor/Non-urgent (Green), and Dead/Unsalvageable/Expectant (Black).

Triage Tags

Use easily recognized triage tags to avoid confusion, to accurately transmit information to other providers, and to avoid duplicated triage efforts. At least initially, premade (formal) triage tags may not be available. Informal tags include marking the patient with the generally accepted triage term, such as "Immediate" or "Delayed" or prominently tying on ribbons that are the triage color (e.g., Black is Expectant or Dead). Paper tags can be secured to the patient's clothes with a clothespin or safety pin and to an extremity with twine or a rubber band. Or write on the patient's forehead (easily seen) with an available marker, such as lipstick or a grease pencil.

An alternative is to use the system the Japanese developed after the sarin subway attack. They use large colored clothespins not only to deal with the standard triage categories, but also to separate patients into those who need "wet" and "dry" decontamination after a chemical exposure. The Simple Triage and Rapid Decontamination of Mass Casualties with Colored Clothes Pegs (STARDOM-CCP) system uses inexpensive (~5 cents each, US) plastic clothespins in the typical triage colors: red, yellow, green, and black, as well as white for patients needing dry decontamination and blue for wet decontamination. Each patient gets a standard (welfare) and decontamination clothespin.

The clothespins are large enough to be easily handled and recognized by responders in protective gear. The tags also survive wet decontamination, although the patients must often hold the pins after their clothes are removed. They recommend having a large sign on-site describing the triage process and colors.38 This type of triage tag can be quickly fashioned from standard clothespins and colored markers or paint.

START (Simple Triage and Rapid Treatment)

This system (Table 7-5) is designed for adult patients in the prehospital setting. Ideally, it allows providers to triage each patient in <30 seconds with the goal of finding the sickest, or "immediate," patients.39 This system (and those who use it) tends to overestimate ("over-triage") the severity of the patients evaluated. The Australians use an almost-identical system that they call Triage Sieve/Triage Sort.40

JumpSTART

For children, use JumpSTART (rather than START) for triage (Table 7-6), since children may normally have a respiratory rate >30/minute, may not follow commands (age and behavioral), and may have a "salvageable period" where they are apneic with a pulse.

SAVE (Secondary Assessment of Victim Endpoint)41

SAVE triage is designed for catastrophic disasters where there are limited transportation and medical resources. Transport may be delayed for days and prolonged when it occurs, and the patient's condition may deteriorate during the interim. This is a secondary (and tertiary) triage system and assumes that START triage has already been done. The system's priorities are based on the following equation:

Value = (Benefit expected ÷ Resources required) × Probability of survival

Patient Classification (in Order of Priority)

• A. Those who will benefit from limited, immediate field intervention. Provide treatment based on two questions:
  • 1. What is the patient's prognosis with minimal treatment?
  • 2. What is the patient's prognosis with treatment using available resources?
• B. Those who will survive whether or not they receive treatment. Provide basic care, as available. Periodically reassess patient status.
• C. Those who will die regardless of treatment. Periodically reassess for improvement.

Patients who are treated should be reclassified if they do not respond. Identify and mark those who would benefit most from transportation to a medical facility (should transportation become available).

Hospital/Health Care Facility

The Israelis use a "continuous" triage method to avoid under-identifying seriously injured patients. They have found that, although experienced triage officers may not be able to initially identify as many as 50% of the victims who suffered life-threatening injuries, the continuous triage system reduced that number to <4%.

The Israeli method incorporates both the decisions made by the triage officers and the primary evaluations performed in the ED. The Israelis divide their ED into three sections according to the severity of injury: (a) mild, (b) moderate, and (c) severe. The decisions made by the triage officer help distribute the victims between the various sections. Once the casualty arrives at the appropriate site, medical personnel quickly perform primary and secondary surveys to identify the severely injured victims. The medical team refrains from using diagnostic resources on those whose exam suggests that they are not seriously injured. If a patient is identified as suffering a life-threatening injury, the trauma surgical team assumes care and transfers the patient to a secondary treatment site. This method has worked well in more than 20 multi-casualty incidents involving >600 patients, missing only 3.8% of those suffering from life-threatening injuries. These patients all had distracting injuries.42

ICU Admission Triage in Disasters

Initial Triage Criteria for ICU Treatment in Disasters/Resource Scarcity

The most useful criteria for deciding which patients should not receive limited intensive care resources come from those developed for use in pandemics (Table 7-7). These patients should then receive standard medical and, when appropriate, palliative care.43–45

When to Abandon Resuscitation

If an attempt has been made to resuscitate a patient, it is pointless to continue when it becomes obvious that there is no hope of recovery. The following signs, if present after 30 minutes of intensive resuscitation, indicate a poor prognosis: fixed and dilated pupils, impalpable femoral and carotid pulses, and no respirations.46

Death Diagnosis with Limited Resources

With limited resources, especially with no access to an ECG or cardiac monitor, clinicians must rely on their clinical exams to determine whether death has occurred—or will quickly and inevitably occur (the "Expectant" [or black] triage category). Without an ECG, mistakes have and will be made. The ECG was the first advance for diagnosing death since biblical times, and it remains the most useful method to accurately determine death.47

Other ways physicians can diagnose death at the bedside are to listen with a stethoscope, feel a carotid or femoral pulse, use an ophthalmoscope to see whether blood in the vessels in the eye has broken into the stationary segments ("boxcars") that signal absence of cardiac activity, and look for cardiac activity with an ultrasound.48,49

Even better are the basic signs used by ambulance crews to determine which patients cannot be resuscitated: rigor mortis, livor mortis (purplish color changes in dependent areas), algor mortis (cooling to room temperature), or an injury incompatible with survival. Note that, when a patient has been exposed to very cold temperatures, severe hypothermia may simulate death. In that situation, warm the patient, if possible, to check for signs of life. Within reason, patients should not be considered dead until they are warm and dead.