Category: Knowledge

Pediatric Infectious Airway Emergencies

Most of us in EMS are pretty good with the typical adult respiratory illnesses — your asthmas, your COPDs, your CHF exacerbations. Maybe the occasional pneumonia. But upper airway illnesses can be harder to keep straight, and two of them are known for occurring especially in pediatrics.

 

Croup

Croup is an infectious disease, usually caused by viral infection of the upper airway. Once upon a time, it was typically caused by diptheria; with the widespread adoption of diptheria vaccination in the US, this has become rare, and the most common cause is now the parainfluenza virus.

Croup is typically seen in children from 6 months to 6 years of age; approximately 15% of all children will contract it at some point.

Its characteristic finding is a coarse, barking, “seal-like” cough. High-pitched stridor is also often present, as is hoarseness of voice, and dyspnea at late stages. All are caused by narrowing of the upper airway due to inflammation from the infection.

This inflammation is normally at the level of the trachea, possibly stretching down into the large bronchi and up into the larynx, but below the level of the glottis.

The onset is usually gradual, lasting from several days to two weeks. Often, there will be fever or coryzal (cold-type) symptoms, such as congestion or headache. A sudden onset is uncommon; usually there will be a day or two of cold symptoms prior to the onset of cough or hoarseness. The upper airway symptoms are often worse at night. (Occasionally, croup may have a spasmodic cause, sometimes triggered by an allergic reaction [akin to asthma]; in these cases coryzal symptoms are rarely present.)

The main differential here should be croup vs. epiglottitis vs. an aspirated foreign body or anaphylaxis. In most cases, these last can be ruled out by a thorough history. As for epiglottitis, it can best be revealed by the severity of symptoms. In almost all cases, croup does not cause significant airway obstruction leading to major dyspnea or hypoxia. A poor general impression (“big sick”), diminished responsiveness, or respiratory failure should point you towards epiglottitis instead. Drooling is not normal. Death is possible but very rare.

Other signs and symptoms are as typical for respiratory difficult in children, including tachypnea, tachycardia, hypotonia (weak or flaccid muscle tone in the exhausted child), chest retractions, and potentially low O2 saturation. In severe cases, as respirations become shallow and weak, stridor and coughing may not be audible.

Treatment is generally supportive. Keep the child comfortable and reassured, if possible with their mother or other caretaker — panic will increase their oxygen demand and may increase stridor. Provide supplemental oxygen as tolerated; blow-by may be most appropriate. Nebulized racemic epinephrine is appropriate field care for significant cases, and most will receive steroids in the hospital. Steam or humidified oxygen is an old standby, and is anecdotally popular, but research suggests it has no benefit.

 

Epiglottitis

Epiglottitis is similar in presentation to croup. Its cause is also infectious, but bacterial instead of viral, and historically the microbial agent was almost always hemophilus influenza type B, a bacterium unrelated to the flu except by name. Effective HIB vaccine became widespread in in the late 1980s; as a result, epiglottitis in the US today is actually more common in adults than in vaccinated children. When it does occur in pediatrics, it is typically between the ages of 3 and 7.

Like croup, epiglottitis is a narrowing of the upper airway. However, it occurs higher, above the glottis, and primarily involving the epiglottis and surrounding tissue. Due to inflammation of this flap at the gate of the airway, a sore throat and difficulty swallowing are almost always present. Fever, stridor, and hoarseness of voice are common, as is a muffled speaking voice and tenderness of the larynx, which may present as severe pain provoked by even gentle external palpation. Typically in children there are no other prodromal symptoms such as congestion; adults often do experience coryzal symptoms.

The chief distinction between epiglottitis and croup is the severity. Although far less common, epiglottitis should be considered an acute emergency, and the narrowing of the airway can lead to significant or total obstruction and resultant respiratory failure. The classical late clinical presentation is the young child found in a “sniffing” position, tripodding or sitting upright, with audible stridor (or quiet respirations as failure approaches), drooling and close to exhaustion. (Drooling results from the inability to easily swallow.) Symptoms can be insidious in onset, but usually progress very rapidly. Mortality in adults with epiglottitis is around 7%, although less than 1% in children.

Treatment may involve supplemental oxygen, but maintaining a patent airway should be your main concern. Intubation may be necessary, sometimes suddenly, so prepare for a difficult airway even if the situation seems relatively controlled. Cricothyrotomy is appropriate as a last resort. Direct manipulation can provoke further inflammation, so the use of basic airways (OPA/NPA) is unhelpful and unwise. Although laryngoscopy can help confirm the diagnosis of epiglottitis (the epiglottis and arytenoids will be cherry red and swollen), it can also provoke spasm and further inflammation, so it is discouraged in the field unless immediate and unhesitating intubation is intended.

Albuterol is not indicated, and epinephrine has no benefit. In-hospital care includes antibiotics to clear the infection, and possibly the use of steroids to manage swelling.

Drug Families: ACE Inhibitors and ARBs

Understanding the renin-angiotensin-aldosterone system is like following one of those dotted-line Family Circus cartoons — not just long and tortuous, but seemingly designed just to be obnoxious.

Here’s the basic idea. The RAAS is the basic system your body uses to control blood pressure, as well as related values like fluid volume and sodium levels. The most important thing to understand is that the activation of this system causes an increase in blood pressure. Following the trail:

First, renin is released by the kidneys. Renin attacks circulating angiotensinogen, turning it into angiotensin I. Angiotensin I is attacked by circulating angiotensin converting enzyme (or ACE), which turns it into angiotensin II. Angiotensin II has various effects, one of which is to stimulate the release of aldosterone.

Whew.

But this isn’t as complicated as it looks. Renin has no real effect. Angiotensinogen just makes angiotensin I. Angiotensin I’s main role is to make angiotensin II. The real money here is in angiotensin II, as well as aldosterone.

Angiotensin II has the primary effect of vasoconstriction. It tightens up the vasculature, increasing blood pressure and systemic resistance. It also produces vasopressin (aka ADH, or anti-diuretic hormone) and aldosterone, which cause the kidneys to downregulate urine production — more fluid will be returned to the circulation rather than discarded into the bladder. Vasopressin also helps angiotensin II to induce further vasoconstriction.

To make a long story short, the activation of the RAAS system causes an increase in blood pressure via both vasoconstriction and a decrease in kidney output. It is always active, playing a key role in maintaining homeostasis; if you sweat out a liter of water running a marathon, or bleed out a liter from a gunshot wound, the system simply upregulates itself to maintain your blood pressure using the remaining volume. (Unlike the sympathetic and parasympathetic systems, which also play a major role in regulating blood pressure, regulation via the RAAS is captained mainly at the kidneys, where low pressure and throughput induces increased renin production.)

Cool? Cool.

Okay, so the role of ACE, or angiotensin-converting enzyme, was to transform the useless angiotensin I into the powerful angiotensin II. What do you think a drug called an ACE inhibitor would do? Indeed: it inhibits the activity of ACE, thus reducing the production of angiotensin II, which then causes reduced production of aldosterone and vasopressin.

Less angiotensin II means less vasoconstriction; the systemic circulation opens up, reducing blood pressure. Less aldosterone means less fluid is retained at the kidneys, so urine output is increased, reducing circulating volume, and again, reducing blood pressure. Handy!

A secondary role of ACE is to degrade, or break down, bradykinin. Bradykinin is basically just another vasodilator. If ACE is inhibited, then less bradykinin will be broken down, hence more bradykinin will be available. The result is more vasodilation — once again, reducing blood pressure.

Readers who can recognize patterns will probably have deduced that ACE inhibitors are used primarily to reduce blood pressure. Obviously, this includes the typical patient with primary hypertension that needs to be managed to reduce long-term morbidity. But it also means other things:

  • Reduced afterload — the resistance the heart has to push against when it pumps blood — means less work for the heart. This is beneficial for patients with heart failure, whose hearts aren’t pumping very well to begin with; or with coronary artery disease, whose hearts need to manage their workload to match the oxygen they’re able to bring in. It reduces “remodelling,” where the heart and the arteries thicken and change shape to better pump hypertensive volumes, with harmful results. And it reduces the damage following myocardial infarction.
  • Reduced preload — the amount of blood that passively fills the heart during diastole — — also means less work for the heart. Greater preload causes greater filling and hence greater contraction, which all means more work, more oxygen demand, and more remodelling. In heart failure, where the heart is unable to fully expel its contents, reduced preload also means less “extra” fluid to back up into the lungs and circulation, and therefore less edema.
  • Better renal function. This is a desirable effect in patients with various forms of kidney disease.

Angiotensin receptor blockers (ARBs) are a closely related family of drugs. Instead of interfering with the conversion of angiotensin I to angiotensin II, they simply prevent angiotensin II from binding with its receptors. The effects are therefore largely the same: vasodilation; reduced aldosterone production, with corresponding greater urine output; and reduction in hypertension with less work for the heart.

The main difference between ACE inhibitors and ARBs goes back to bradykinin. If you remember, ACE plays two roles: converting angiotensin I to angiotensin II, and degrading bradykinin, a vasodilator. Since ARBs have no effect on ACE, bradykinin is broken down normally. This may result in slightly less vasodilation, but it also reduces the side effects of elevated bradykinin, which can include edema and a nasty cough. ARBs are most often used in patients who can’t tolerate ACE inhibitors.

Overdose on ACE inhibitors is generally unremarkable. The main effect is hypotension, but it is rarely severe.

 

Examples

Generic names of ACE inhibitors and ARBs are very, very easy. Trade names are harder, but do have some common elements.

ACE inhibitors

  • Drugs ending in -pril are invariably ACE inhibitors (enalapril, ramipril, captopril, lisinopril, etc.)
  • Drugs ending in -ace are often trade names of ACE inhibitors (Altace, Tritace)
  • Drugs ending in -tec are often trade names of ACE inhibitors (Vasotec, Renitec, Novatec)

ARBs

  • Drugs ending in -sartan are invariably ARBs (losartan, valsartan, candesartan, etc.)

More Drug Families: Stimulants and Depressants; Steroids and Antibiotics; Anticoagulants and Antiplatelets

What it Looks Like: Seizure

See also what Agonal Respirations, Jugular Venous Distention, and Cardiac Arrest and CPR look like

A seizure is an episode of chaotic, disorderly electrical activity involving part or all of the brain. It is most often seen in epilepsy, but seizure can also occur acutely due to hypoglycemia, eclampsia, stroke, head trauma, alcohol withdrawal, and other causes.

Seizures are typically divided into two major types, partial seizures which involve only a portion of the brain, and generalized seizures which involve the entire brain.

Partial seizures are further divided into simple partial and complex partial seizures. In a simple partial seizure, consciousness is maintained, but unusual sensory, motor, or emotional sensations are observed — muscular tics, visual disturbances, strange feelings, and more are all possible depending on the area of the brain affected. Most often, this will then proceed into a larger seizure, in which case these early effects are called an aura, and used as a warning sign. Complex partial seizures are similar, but involve both hemispheres of the brain, and are distinguished by a loss of awareness or memory — the individual’s consciousness is impaired during the episode. This is the most common form of seizure.

The best known generalized seizures are tonic-clonic seizures, known historically (and still called by many laymen) “grand mal” seizures. They are characterized by two phases: a tonic phase, where the body becomes rigid and immobile, followed by a clonic phase, where full-body involuntary muscular jerking occurs. This is usually followed by a post-ictal period, where the patient may be unresponsive, or behave unusually, appearing combative, stuporous, or otherwise impaired. Either the tonic or clonic phase may be minimal or absent.

Absence seizures, historically “petit mal,” are characterized by a loss of awareness with a lack of outward activity. The individual may simply stare without moving or speaking, and after cessation of the seizure resume where he left off with no memory of the episode. Absence seizures may also present with some outward seizure activity, in which case the distinction between types becomes blurred.

Febrile seizures are seizures caused by elevated temperature (usually >100 degrees), most often seen in infants and young children. They are typically tonic-clonic in nature and almost always have benign outcomes; they rarely go on to develop into adult epilepsy.

Status epilepticus describes a prolonged seizure state, customarily defined as a seizure lasting over 30 minutes or multiple seizures without a full recovery in between. Some authorities draw the line at any seizure over 10 minutes, and there is evidence that even seizures longer than 5 minutes are unlikely to end without medical intervention. Status epilepticus is a true life-threatening emergency with high mortality; the continued chaotic activity of the brain can lead to permanent brain damage or death. Definitive treatment is the use of anti-convulsants, which attenuate the neuronal activity; in the field these are typically benzodiazepines like lorazepam (Ativan), diazepam (Valium), or midazolam (Versed). Since the duration from 911 call to EMS arrival on scene is often greater than 5-10 minutes, a seizure that is still ongoing upon your arrival should raise immediate suspicion of status epilepticus; a careful history should be obtained from bystanders when possible, including time since onset and any intervening recovery.

In some cases, seizures will be followed by a persistent, unilateral focal weakness in muscles that were active during the seizure. This is called Todd’s paresis, and since it can closely mimic the signs of stroke (even impairing eyesight or speech), it is wise to ask about recent seizure activity in patients with a history of a seizure disorder who present with signs of stroke.

Field care for seizure generally involves preventing secondary injury, such as blunt trauma caused by hitting or landing on nearby objects. During the tonic phase, respirations may be minimal, resulting in cyanosis; this is usually brief enough not to cause harm. The greatest concern is to maintain an open airway and prevent aspiration; when possible the patient should be placed in the lateral recovery position to help prevent soft tissue obstruction and allow fluids to drain away. Suction may be valuable, and an NPA may be considered in prolonged episodes. Supplemental oxygen is always appropriate, although a non-rebreather mask may not be tolerated in the post-ictal period. If respiration appears inadequate in prolonged seizures, positive pressure ventilation (by BVM or invasive airway) may be attempted.

This video from Dr. Robert S. Fisher is an excellent summary of the basic types of seizure. (Here is another on partial seizures; these are unusually good educational videos for a free resource.)

Here is an example of a simple partial seizure in a child, in this case manifesting as a repetitive facial tic. Note that the child retains consciousness throughout.

Here is an example of a complex partial seizure, also in a child. Note the repetitive, aimless movements of the arm and head, which are known as automatisms and are wholly involuntary; if spoken to, she would not respond.

Another complex partial seizure, in a young adult. Note the automatisms of the mouth and the wandering posturing of the arm.

An absence seizure in a child. Note the lack of any outward signs, except a total lack of responsiveness.

An excellent video of a tonic-clonic seizure in an adult. Note the labored breathing and obvious altered level of consciousness post-ictally.

Another good tonic-clonic in an adult. You see his awareness of its onset due to an aura, followed by gradual tonicity and then clonic jerks. Also note the snoring respirations; better positioning (and the suction catheter that the nurse couldn’t find) would have helped here.

Tonic-clonic in an infant, this one of febrile etiology.

Tonic-clonic in a sleeping adult; skip to 1:00 if you see better with lights.

Live from Prospect St: Dizzy at Hillcrest (part 3)

Continued from Part 2

My apologies for the delay on this update: there have been major computer troubles here at EMSB HQ. We’re back in action now with the final piece of our scenario.

Ultimately, this patient was rapidly packaged and transported emergently to the nearer facility for immediate imaging to rule out intracranial hemorrhage. Her final diagnosis and disposition are not known.

This case demonstrates the ambiguity we’re often faced with in the field, where we may encounter findings in our assessment that are suggestive of Badness, but not definitively so. Particularly when faced with a patient whose complaints are minor or who generally presents well, it can be difficult to make the call to upgrade these patients to a higher level of care. Nobody wants to be the Boy Who Cried Wolf. However, our job is to get people to the most appropriate care, and although we should try to minimize overtriage, within reason, safe is better than sorry. The situation can be particularly difficult when we are dispatched as a low priority to an unremarkable complaint; changing gears from a low- to a high-severity mode takes more balls than merely continuing what’s already been set in motion.

 

Assessment: The Pink Flags

The suggestive if not outright alarming findings (I like to call them “pink flags” — not quite red, but close) with Ms. Smith were the following:

  • A recent fall, reportedly with a blow to the head and loss of consciousness.
  • A subsequent (apparently new) complaint of dysnomia (the inability to express oneself in words, a form of aphasia), which suggests some sort of neurological or metabolic insult.
  • A subsequent and sudden onset of vomiting with no other apparent explanation. This could be a sign of hemorrhagic stroke, although more minor head injuries can also induce vomiting.
  • A history of Coumadin (warfarin) use — a “blood thinner” or anticoagulant — which is a risk factor for intracranial bleeding.
  • A complaint of “head pressure,” which remotely suggests headache, typical in head bleeds.
  • A reported positive finding on a neurological test (failed finger-to-nose), which potentially supports a neurological event.
  • A complaint of dizziness, which is suggestive of either a balance-type (inner ear) pathology or a neurological one.
  • A finding of hypertension, which may or may not be elevated above the patient’s baseline.

On the other hand, the following findings point generally away from the likelihood of a stroke or intracranial bleed:

  • An alert and oriented patient mentating at her cognitive baseline.
  • A normal Cincinatti Stroke Scale, which assesses for arm drift, facial droop, and speech slurring.
  • A lack of other “focal” neurological deficits (an abnormality that is localized to a single sensory or motor region, such as a droop in one half of the face, or loss of sensation in the left arm but not the right). She has equal peripheral CSM, no complaints of partial vision loss, and so forth.
  • A lack of any significant headache. Although there is a vague complaint of pressure, which could be explained by the actual trauma to the head, headache associated with intracranial hemorrhage is typically severe and sudden.
  • Equal and non-dilated pupils. (Although they do present as small, this is an unremarkable finding in the elderly, as is poor reactivity — constricted pupils can’t constrict much more.) Furthermore, the eyes track well towards all sectors; gaze paralysis is suggestive of brain damage. None of this is highly predictive, however.
  • A lack of rigidity of the neck, which would support a hemorrhage.

Taken together, this cloud of positive and negative findings produces our clinical picture. We are not so fortunate that any one finding is diagnostic, or highly suggestive to either rule in or rule out Badness. Rather, we have a constellation of weak findings.

 

Differential: Strokes and Bleeds

It can be important to make a distinction between intracranial hemorrhage and stroke. Intracranial hemorrhage (we’ll call it ICH, not to be confused with “intracerebral hemorrhage,” discussed below — both abbreviations are seen in the literature) describes bleeding anywhere inside the dome of the skull, typically from a ruptured vein or artery. Sometimes, this occurs inside the skull but outside the brain, between the various membranes that lay between brain and skull: epidural (outside the dura), subdural (inside the dura), and subarachnoid (inside the arachnoid) are the main types and locations.

Bleeding deep within the tissue of the brain itself is also possible, and is a subcategory of ICH called intracerebral hemorrhage.

A stroke is a localized injury to brain tissue resulting in permanent neurological deficits. By far, the most common cause is known confusingly as ischemic stroke, and describes an event where a clot or other obstruction blocks an artery that feeds a portion of the brain. (This is the same mechanism that damages the heart in a myocardial infarction.) The other main cause of stroke is hemorrhagic, when an artery bleeds openly into the brain, causing damage both from the loss of perfusion to downstream tissue, as well as from the pressure caused by the growing pocket of blood. This is where stroke and head bleeds intersect: when either an intracerebral or subarachnoid hemorrhage is sufficient to cause local neurological damage and permanent loss of functional brain tissue, a stroke results. Epidural and subdural bleeds do not cause stroke per se, although they can still result in acute neurological symptoms due to the increase in intracranial pressure.

Although the effects of stroke are similar with either ischemic or hemorrhagic etiologies, hemorrhagic strokes may additionally produce the telltale signs of rising intracranial pressure, such as headache, vomiting, general (non-focal) neurological deficits, and in the late stages, Cushing’s triad (bradycardia, irregular respirations, and hypertension).

 

Applying the Differential

Ms. Smith’s history is certainly suggestive for a bleed. Head trauma is the most common cause of ICH, and with her Coumadin use, she should probably be worked up regardless of her minimal complaints. Her additional neurological complaints make this a potential “uh oh,” advising transport to a facility that can provide immediate care. However, there are some notable negatives that tamper this enthusiasm.

For one thing, it would be unusual for a bleed of this type to present so inconspicuously. If severe, we would expect to see a profoundly altered mental status, up to and including outright coma, and probably a significant headache. If there is also the localized infarct of a stroke, we would expect focal neurological complaints — local damage should cause focal deficits. The reason that the Cincinatti Stroke Scale uses facial droop and arm drift to screen for stroke is because the majority of strokes will be revealed by unilateral deficits. Ms. Smith has none of this.

If there is indeed a stroke, the type most consistent with her presentation is probably a cerebellar stroke affecting the vestibular (balance) system. This region is responsible for coordinating motor and sensory signals, allowing synchronized behavior, such as the finger-to-nose test she failed. It’s also responsible for proprioception and balance; hence, damage could produce her complaint of dizziness. It is always important to distinguish “dizziness” (a sensation of spinning, consistent with either vestibular stroke or BPPV) with “lightheadedness” (a dimming of the vision, as seen in orthostatic hypotension). This is a notable possibility mainly because cerebellar injuries often do not produce the focal deficits characteristic of other strokes.

If you are very enterprising, Dr. Scott Weingart describes a three-test screen (introduced by Dr. David Newman-Toker and Dr. Jorge Kattah here) which can help catch vestibular stroke in borderline cases such as these. It uses two simple and easy tests, plus a third — involving a head twist — which is more difficult to assess and vaguely terrifying to perform. If you plan to use any of them, it’s the sort of thing you should be practicing beforehand. (I personally find the head twist finicky and liability-prone in most circumstances.) Like all such tests, their role in the field should only be to help determine transport destination and priority, and give you additional information on how hard to push a reluctant patient towards transport. It is not appropriate for enterprising Dr. Medics to use as ammunition to say, “oh, it’s negative, you’re clearly fine.” The weight of a thousand lawyers will descend upon you, and rightly so, the day you decide that you have the power to rule out major sickness from your ambulance.

If an extra-cerebral hemorrhage proved to be the culprit, a subdural bleed is probably the most plausible, due to the relatively slow and insidious development of the symptoms.

Additional tests that were not performed, but might have been useful, include a visual field test (testing at minimum eyesight in both visual hemispheres), a “stick out your tongue” test (looking for deviation to either side), and a more complete test of reasoning and recall (portions of the Folstein Mini-Mental, for instance).

Many of the major components of the peripheral neurological exam we performed are taken from this excellent lecture by Dr. Gene Hern of AMR Contra Costa County (see 37:20 through 40:50), and is my favorite expansion on the typical “squeeze my hands.” Sharp sensation can be tested with the tip of a pen — or you can use Dr. Hern’s pinching method.

Two other tips: when performing the facial droop test, “show me your teeth” produces better results than “smile” — patients tend to give a larger, more symmetrical smile using more muscles. And when testing for arm drift, remember that the patient’s eyes should be shut, and the hands should be facing upward (supinated); this is a more difficult test and therefore more sensitive.

 

Treatment and Transport

The key points on our differential therefore come down to two: intracranial hemorrhage vs. anything else. “Anything else” could be any number of things that produce diffuse and global symptoms, including metabolic problems or even a brain tumor. Diabetic etiologies are always be a possibility, although glucometry was fortunately available to rule that out. In general, the old standby AEIOUTIPS is the sort of thing we’re looking at here. And remember, multiple concomitant pathologies are just as likely as one all-encompassing Badness, if not more so. As a starting point, we should bear in mind that around two-thirds of falls with loss of consciousness in the elderly will end in death. The risk is high.

As always, the differential only matters to the extent that it will affect our decisions. What will our field treatment be?

Certainly oxygen. Although hypoxia is unlikely to be significantly contributing to Ms. Smith’s complaints, it could be playing a role. Depending on local protocol, low-flow through a nasal cannula may be plenty.

In the case of stroke, there is some evidence that hyperoxygenation with high-flow O2 can contribute to worse outcomes. The 2010 Emergency Cardiovascular Care guidelines from the American Heart Association recommends titrating oxygen therapy to maintain an oxygen saturation of at least 94%, but not necessarily slapping on a non-rebreather at 15LPM. Depending on whether oximetry is available to you, and depending on your local policies and attitudes, this may or may not fly; it’s something to ask your boss and medical director.

What about C-spine immobilization? As always, this will be a matter of opinion and protocol. In some areas, any fall from standing height, with a blow to the head — especially for an elderly patient — must always be immobilized. However, clinically I would not consider it indicated here. Whatever criteria or standards you adhere to for selective immobilization, Ms. Smith likely meets them: she has had no peripheral neurological deficits (weakness, tingling, numbness, pain), no neck or back pain or tenderness, no factors that would impair her reporting of the above (such as distracting injuries or altered mental status), turns her head freely, and although not ambulatory on our arrival was obviously ambulatory for several hours prior. Remember that the only reason for the immobilization of blunt head trauma patients is the suspicion that any injury substantial enough to cause ICH may also be substantial enough to cause a cervical spine fracture — and while a valid reason for suspicion, this is just one factor to consider. (Conversely, if we had found focal neurological deficits, we would have likely been unable to determine whether it was secondary to the suspected ICH, or secondary to a spinal injury — immobilization would have been unavoidable.)

Close monitoring will be warranted, especially if we do suspect a bleed. Although Ms. Smith appears currently stable, there is a real possibility of her mental status deteriorating; epidural bleeds in particular are famous for a “lucid interval” following the initial trauma, after which the patient suddenly and catastrophically decompensates. Control of the airway and ventilatory support should be provided as necessary. If there are signs of herniation syndrome — an acute rise in intracranial pressure, resulting in “coning,” or the brain being forced through the openings in the skull — it may be reasonable to hyperventilate the patient slightly, at a rate of 1 breath every 3 seconds. Although the drop in systemic CO2 caused by a higher ventilatory rate results in a systemic respiratory alkalosis (high PH), which tends to reduce inflammation and hence lower intracranial pressure, it also reduces cerebral perfusion; it is therefore no longer recommended as a routine practice. Intracranial pressure is a challenging problem that produces a physiological tightrope that we need to delicately walk; hyperventilation is a last-ditch flailing that’s only advisable when things can’t get much worse.

Is an ALS intercept appropriate? Again, this may depend on your protocols. As Ms. Smith currently presents, there is no benefit to ALS care; whether or not she’s hemorrhaging, that’s a matter for the hospital, not the field. However, if should deteriorate, then ALS could prove very valuable in the management of her airway, seizures, cardiac arrhythmias, and other complications. With Ms. Smith’s currently excellent clinical picture, and the short transport to definitive care, I would not attempt to meet the paramedics unless I tripped over them in the driveway. However, the opposing argument can easily be made, and I wouldn’t call it wrong.

The most appropriate destination for this patient will likely be the nearest primary stroke center. A “primary” stroke center is required to have various resources available 24/7, the most important in our case being a CT scanner. The definitive determination of the presence or absence of our possible bleed will be via some form of CT, or possibly by MRI (if available).

Treatment may or may not involve surgical intervention, depending on location and severity. Many of these cases are managed conservatively, both because the benefits of surgery are often small and the harm (especially in deep brain bleeds) often large. As a result, my personal inclination is to steer towards the nearest facility that can provide immediate imaging; if surgical intervention beyond their capabilities is found to be indicated, transfer can be arranged. I would not advise transporting to the more distant requested facility; the only notable benefit other than the patient’s convenience and comfort (which we won’t diminish) is that her medical records and following physicians may be available there, and her history doesn’t seem complex enough for this to matter significantly.

In some areas, a few hospitals are designated as “comprehensive” stroke centers, a step above primary. These facilities are specialty centers with the most advanced stroke management capabilities, which may include diagnostic and interventional methods that would be appropriate to us. The system of comprehensive centers is still inchoate and only available in some states; check if yours is one of them.

Your local hospitals may follow a prehospital protocol that allows for a “stroke activation,” similar in principle to trauma or cath lab activations, where appropriate resources are mobilized by request of EMS and waiting upon your arrival. Depending on the local indications (for instance, your hospitals may demand a positive Cincinatti Stroke Scale), Ms. Smith might qualify.

 

Conclusion

In the end, I was unable to obtain patient follow-up on Ms. Smith. She received low-flow O2, was not C-spine immobilized, and was diverted to the nearer stroke center with an emergent transport and no ALS. An entry notification was made with an advisory of her status, although no formal stroke alert was given. She was stable throughout.

It’s important to note that our assessment of Ms. Smith, our analysis of her differential, and our resulting treatment and transport decisions, are not actually dependent on her eventual diagnosis. It doesn’t matter whether we ended up being “right” — hence, it doesn’t matter that we never found out the “answer,” even though I do love a good puzzle and I admit that I wanted to know. As long as we made an appropriate interpretation of our assessment findings, and made appropriate decisions based on them, then we got it right. Perhaps her complaints turned out to result from an alien egg incubating in her chest; that wouldn’t make us wrong, it would only mean that she was an aberration. Our business in the field is to play the odds in a responsible way, weighing risk-vs-benefit to provide our patient with the best chance of a good outcome.

That’s all. And that’s plenty.

Live from Prospect St: Dizzy at Hillcrest (part 2)

Continued from Part 1

While you chat, your partner helpfully places Ms. Smith on a nasal cannula running oxygen at 4 LPM.

You ask whether she lost consciousness when she fell, and she agrees that she may have briefly. When you ask why she fell, she states she simply tripped on the fringe of the rug. The fall was 3 hours ago, and she first vomited immediately afterwards. Until this morning, she was feeling normal, with nothing bothering her over the previous days. Her chief complaint seems to be her persistent inability to “find words,” although staff state that they called EMS mainly due to her dizziness.

When you pursue the “pressure” behind her eyes, she admits that it’s a pain of sorts, but it is obviously not too severe, and she refuses to quantify it with a number. She clarifies her dizziness by stating that although it may be worse when she stands or walks, it is continuous; she is experiencing it even as you speak.

Her pupils are equal, somewhat small, and react slightly to light. Her eyes track in all directions, with no appreciable nystagmus. When you ask her to show her teeth, she does so with no facial droop. When you ask her to hold her arms straight in front of her, palms up, with her eyes closed, she does so with no unilateral drift. She demonstrates good, equal grips, equal bilateral strength in finger-abduction and wrist flexion/extension, and equal bilateral strength in ankle dorsiflexion/plantarflexion. Her radial pulses are equal, as are her dorsalis pedis pulses, and she notes normal bilateral sensation when you pinch her hands and feet.

Throughout your conversation, she has demonstrated no slurring of speech, normal recall, and excellent orientation.

Consulting with the staff, you learn that her medications include Metoprolol, Simvastatin, Metformin, Lisinopril, Colace, Aspirin, and Coumadin for a recent hip surgery. She is allergic to Penicillin. They are unsure about her baseline BP, but Ms. Smith believes it is normally “in the 140s.” Staff believe her temperature has been recently normal, although they aren’t certain.

Your partner obtains her blood glucose at 149.

The nearest ALS is 15 minutes away.

At this point, what are the leading possibilities in your differential?

With that in mind, what is this patient’s priority?

What is your transport destination?

Is any treatment needed at this time?

Should you make any calls to mobilize further resources?

Live from Prospect St: Dizzy at Hillcrest (part 1)

Many moons ago, there was an enlightened discussion list run by Jeff Brosius and Valerie DeFrance called Live from Peachtree St. The format was this: Jeff would periodically present scenarios based on real-life calls, putting them out for debate on diagnosis and treatment; later he’d provide the outcomes, with a discussion of the relevant clinical issues. It was a great educational model, and I’ve always thought that scenario-based instruction was a fantastic way to learn to parse the details of a call, determine what matters, come up with a working diagnosis, and make your decisions. You can’t learn hands-on skills from scenarios, but you can exercise your noodle, and more noodle-exercise is what this business needs.

Sadly, Live from Peachtree St. closed its doors years ago, although its archives are available online and still a great read. In the same spirit, however, and with permission from the authors, I’d like to revive the tradition. With that in mind, and harkening from my own neck of the woods, I give you: Live from Prospect St!

To juggle the format a little, I’d like to break these cases up. Remember “Choose your Own Adventure” books? In a similar fashion, I’ll give you the first chunk of the call, let you analyze the facts and make some decisions, then go forward with another round of details, and finally present the outcome. Cases are either true to life (obviously edited for HIPAA) or closely based on reality with editorializing for educational purposes. Feel free to send in any cases of your own! We’re not only interested in intriguing or unusual calls, but also more run-of-the-mill cases that illustrate important fundamentals of assessment and care.

All levels are invited to play, but in keeping with our mission, we will assume a BLS scope of practice. Onward!

 

Dizzy at Hillcrest

You are the technician on A-8, a dual-EMT, transporting BLS ambulance. You are equipped with BLS epinephrine, aspirin, glucose, nebulized albuterol, pulse oximetry, and finger-stick glucometry. Intercepting ALS is available upon request. It is noon and the weather is fair.

The radio crackles: “Ambulance 8, respond cold to Hillcrest Manor, in the lobby, for the patient with dizziness.”

Driving non-emergently, you arrive several minutes later at an assisted living facility. In the lobby, you are met by staff, who direct you toward an elderly woman in a chair. A nurse informs you she has been feeling somewhat unwell all day, and 30 minutes ago began complaining of dizziness. She also states that she failed a finger-to-nose neurological test. They request that you transport her to Mount Doom Hospital, a high-quality community hospital several towns away (20 minute transport routinely; 13 with lights and sirens); it is a cath lab and stroke center, although not the nearest facility for either, and is not a trauma center. The nearest facility is a community hospital of similar size and capabilities (10 minutes routinely; 5 with lights and sirens).

You kneel in front of the patient, noting that she is alert and appears generally well, in no obvious distress. She introduces herself as Ms. Smith, is 68 years old and fully oriented, and tells this story:

“This morning when I was getting out of bed, I fell and bumped my head. After I got back up, I was trying to do my writing assignment for our group, but I kept finding that I just couldn’t seem to think of the words — I’d stare at a sentence for ten minutes without knowing how to finish it. I’ve been feeling a little dizzy most of the day, and a little sick — I vomited a few times. And I feel like there’s a pressure behind my eyes.”

While you talk, your partner obtains these vitals:

Skin: dry; unremarkable at the core; cool and slightly pale in the extremities
Pulse: 90 [at the radial, strong and regular]
Respirations: 14 [regular, normal depth and unlabored]
Blood pressure: 164/98
Oxygen saturation: 96%

When asked, staff report that Ms. Smith’s medical history includes coronary artery disease, diabetes, mild dementia, and hypertension.

What is your current impression of the patient’s priority?

What are the leading possibilities in your differential?

What further assessments or information would you like to better inform your decisions?

What actions or interventions, if any, would you like to take at this time?

Post responses to the comments.

Managing STEMI Mimics in the Prehospital Environment

Here’s one for the medics in the audience, or anyone interested in the box with the squiggly lines.

ST elevation means acute MI. Or does it? Most medics understand that this isn’t always the case, but many don’t recognize how often it’s not, and looking deeper — sorting out true STEMI from the many non-MI pathologies that also produce ST-elevation — is not the easiest task.

The following is a PowerPoint presentation I produced for use in either continuing education coursework or merely as a standalone reference. The main intended audience is EMS providers at the paramedic level, but most of the information is at least somewhat relevant to all levels of care.

The topic is the recognition and management of “STEMI mimics,” to steal Tom Bouthillet‘s phrase: non-ACS conditions that nevertheless present with signs and symptoms resembling acute MI, particularly ECG changes such as ST segment elevation.

This is best taught by a knowledgeable instructor, but it’s designed to be usable as a self-contained reference for ambitious students; these are info-rich slides, not just graphical accompaniments to a lecture. It does assume a foundational paramedic-level education, as well as a basic understanding of ideas like sensitivity and specificity — a review of our tutorial may be in order. The illustrative ECGs are labeled with “answers” in the slide notes of the PowerPoint versions, although not on the PDF, so that’s probably the best version if this material is new to you.

Although fairly comprehensive, it’s intended as a practical guide for field assessment and treatment, rather than an in-depth examination of the etiology and course of care for every pathology discussed. For additional information, the sources for most of the contentious claims and data are listed on the slides; sources for the more everyday material are available by request. And remember to follow your service’s protocols and understand exactly where and how far you have flexibility to make some of these calls; in many cases, the decisions will be made for you.

There are 192 slides in the full presentation; the most common feedback is that this can make for a very dense and potentially drawn-out class. There is one natural “intermission” point for a break at about the halfway mark, between the introductory discussion on general ideas and before diving into specific and individual mimics. If desired, the course can be broken up further into multiple units or even multiple days.

Feel free to share, redistribute, or use for your own purposes; this is educational material made available without charge or obligation.

[Edit 10/28/12] This presentation was later enhanced into a narrated video lecture

Reading Research: Outcomes

Some folks say that our business is “saving lives.”

Sometimes they’re kidding, sometimes not. Either way, the problem with this kind of attitude is simple: whenever you’re not saving lives, you won’t feel like you’re doing your job. And it takes about ten seconds into an EMS career before you discover that most of what we do isn’t saving lives.

Because most of our patients aren’t in danger of dying. And many of the ones who actually are will die in days, weeks, or years — which doesn’t mean that our care doesn’t matter, but it’s hard to call it a “save” when they were going to survive your time together no matter what you did. That’s just not what’s going on.

I realised that my training was useful in less than ten percent of the calls, and saving lives was rarer than that. After a while, I grew to understand that my role was less about saving lives than about bearing witness. I was a grief mop. It was enough that I simply turned up. (Bringing Out the Dead)

In the field, we usually understand this fact. But for some reason, when we conduct research, we forget. Read ten studies on prehospital medicine, where interventions, methods, or systems are tested, and in nine of them, the primary outcome examined will be mortality. Did more patients survive than otherwise would have? Did the thing we’re examining save any lives?

Obviously, this is an important consideration. Saving lives is good! But as we’ve already agreed, most of our care isn’t lifesaving, no matter what happens. Lives are simply not at stake. Asking about mortality improvements in ankle fracture patients is like asking how we can reduce the number of car thefts on the moon.

If someone’s not calling us because they’re dying, what’s their problem? Usually, they’re ill or injured, but not so badly that their life is endangered. They still need medical care, but its purpose will be to reduce their long-term morbidity, as well as their short-term pain and discomfort. It will be to maintain the optimal function of their limbs, the greatest cardiac output, the most neurological capacity — to make sure the person who walks out of the hospital can still do most of the things he wants to do, needs the fewest meds, is fated for the fewest procedures, and experiences the least chronic agony.

And when someone’s lying in front of us, moaning from a broken femur, or turning green from nausea, we can help with that too. Many, many of our patients have a chief complaint involving pain or some sort of discomfort, and although we typically look at it as a diagnostic sign suggesting an underlying problem — which it is — odds are good that the patient is more interested in feeling better than in a diagnosis. We can help with that (at least, the medics can). Hell, even us BLS fellas can relieve some stress, grief, and psychological distress with a calming demeanor and sympathetic word.

The fact is, it’s shortsighted and wrong to only see mortality when we’re talking about our work, and it leaves out a huge amount of what we do. If we never did anything but relieve terrible pain, and never saved a single life, we’d still be performing a wonderful service. Our mortality myopia has gotten so bad that many providers need to be told that pain is associated with greater morbidity and mortality (which it is) before they’ll even consider managing it as a priority! Pain is the ultimate subjective experience; when it’s yours, nothing could be worse, but when it’s someone else’s, it’s hard to give two craps. As Dr. Jeff Guy likes to quote,

We must all die. But that I can save him from days of torture, that is what I feel is my great and ever-new privilege. Pain is a more terrible lord of mankind than even death itself. (Albert Schweizer)

What this all means when it comes to medical research is you’re going to see many, many studies that determine a certain intervention “no better than placebo,” when what they really mean is it didn’t save anybody’s life. But that doesn’t mean it didn’t help anybody. There could have been profound effects on short- and long-term morbidity, and you could have made someone far more comfortable and happy. That won’t be reflected in those numbers. The fact is, it’s difficult to study these things, because they’re harder to quantify than mortality, and they seem less important (perhaps they are), so you just don’t see as much research that seriously examines such outcomes.

So although I’m a strong proponent of evidence-based medicine and turning to research to further our standard of care, the answers won’t always be there. Partly this is the fault of the folks who do the studies, and partly it’s an inherent limitation to the scientific method. Either way, it means that the literature needs to be supplemented with our personal experience and the experience of our colleagues, and even then, we may not have the whole picture. It’s a drag, but that’s how it is.

Personally, I like to see my job as first to reduce acute suffering, second to manage permanent or long-term morbidity, and lastly to reduce mortality. This is a ranking born of frequency, not of urgency; obviously, if someone’s actually dying on me, saving their life takes precedence. But it’s no sweat off my back if they’re not, because there’s still a great deal more to be done. And it’s a shame that the research can’t always help get us there.

More on Reading Research: Diagnostics

What it Looks Like: Agonal Respirations

See also what Jugular Venous Distention, Seizures, and Cardiac Arrest and CPR look like

Education and experience are both important to making a well-rounded provider, and each of the two have distinct advantages. Perhaps the greatest advantage of experience is that it gives you the best ability to recognize situations you’d otherwise only know by description or by photograph.

Nowadays, though, with the Wonders of Modern Technology, we have some tools that can help bridge this gap. Experience is still essential — but there’s no reason that the first time you see a seizure or cyanosis should be in a situation with real stakes.

So let’s go through some of the common medical events and conditions we talk about, learn about, but may not truly know the presentation of until we encounter it.

Today, it’s:

 

Agonal Respirations

Agonal respirations are an inadequate pattern of breathing associated with extreme physiological distress, particularly periarrest states (that is, it is usually seen just prior to cardiac arrest, as well as during and for some time after). Although not always seen during arrest, it is not uncommon, and there is some evidence that it may be associated with better outcomes than arrests without agonal breathing. Whatever the case, it can easily be confused for ordinary respiration, leading to the mistaken impression that the “breathing” patient must also have a pulse; this confusion is part of why the American Heart Association no longer recommends checking for breathing as part of layperson’s CPR.

As for healthcare providers, whether we’re able to put the label of “agonal” on it or not, we should be able to recognize from the rate and depth that this is not adequate respiration to sustain oxygenation, and ventilatory assistance (as well as a check of hemodynamic status) is in order. But recognizing the specific nature of this breathing can be a very useful red flag to set your “code” wheels in motion.

Here are a few simulated examples, performed by medical actors. They range in presentation and context.

http://www.youtube.com/watch?v=M99bRjmsC7E

Finally, here’s a treat — this is a video of a real-life cardiac arrest at a beach in Australia. Starting after the first shock, from 2:39 onward, you can see a great example of agonal breathing. The rest of the video is also a nice example of an honest code being worked in the field — not perfect, but real. (For bonus points, how could their CPR and other treatment have been improved?)

(Thanks to Dave Hiltz for inspiring today’s topic.)

Reading Research: Diagnostics

Why do you do what you do when you do it?

Initially, we do it because we’re taught to. It comes down to us from on high, from instructors or textbooks — when you see x, do y — and we’re simply expected to learn it, memorize it, and recite it back. Then, once in the field, to follow it mechanically.

But before long, if we’re to become more than just medical Roombas, we really have to start asking Why. It’s not because we’re difficult children, or to satiate our curiosity. It’s because even at the best of times, the rules can’t address every situation. And in order to make intelligent, appropriate decisions when the circumstances aren’t clear and simple, we need to understand the underlying principles behind the rules we learn. We need to understand both the potential value and potential harm of the interventions we provide. We need to understand the meaning and importance of specific assessment findings. We need to be students of reality, and the human body, rather than of arbitrary rules.

In order to do all this, we need to be able to read research. Medical research is where these answers come from; it’s where we learn what works, and how well, and what importance to attach to the things we see. To read research, though, we need to understand the basic statistical methods they use.

Statistics is a big, big topic, and I don’t have a strong background in it, so if you really want to dive into this, take a class. The analytical and regressive methods used to crunch the data in a study are something we won’t touch here. But we do need to understand a few basic terms, because they’re central to how the results of a study are presented — in other words, if you’re looking for answers, this is the language in which they’re written. So although the idea of a post about statistics may sound as appealing as a brochure on anal ointment, bear with me; this won’t be too painful, and it’s information you can use over and over and take to your grave. Right now, let’s talk about numbers used to describe accuracy of diagnostic signs.

 

Sensitivity and Specificity

Take a certain test. It could be anything. A clinical finding. A laboratory test. Even a suggestive element from a patient history. Call it Test X.

Let’s say that this test is linked to a certain patient condition, Condition Y. Something bad. Something we want to find. In fact, Condition Y is the whole reason we’re looking at Test X.

What would make Test X a good test for Condition Y? Well, when the test says “You have Condition Y!”, then you should really have it. And if it says “You don’t have Condition Y!”, then indeed, you shouldn’t have it. It doesn’t have to be perfect. But it should be pretty good — otherwise, what’s the point in using the test? If it doesn’t tell us something we didn’t know before, we might as well ignore it.

When the test says “You have Condition Y,” and you really do have it, we’ll call that a true positive. When the test says, “You don’t have Condition Y,” and indeed you don’t have it, we’ll call that a true negative. Those are the findings we want; we want the test to tell us the truth, so we can base our treatments and decisions on reality.

On the other hand, when the test says, “You have Condition Y,” but you DON’T have it — in other words, an error, the test got it wrong — we call that a false positive. We thought you were positive, but whoops, you’re actually fine. And when the test says, “You don’t have Condition Y,” but it turns out that you do, we’ll call that a false negative, or a miss. The test cleared you, but it missed the badness; you actually do have the condition. These are the screw-ups.

How many true positives and true negatives does our test yield, versus how many false positives and false negatives? This determines how good our test is, how faithful to reality. The perfect test would have 100% true results, either positive or negative depending on the patient’s condition: if you have Condition Y, the test is positive, and if you don’t have Condition Y, the test is negative. There would be zero false positives or false negatives.

The worst possible test would have about 50% true and 50% false results. There would be no correlation between the test results and having the condition. In fact, it would be pointless to call this a test for Condition Y; we might as well flip a coin and call that Test X, because it would be just as useful.

Okay, so how do we determine the accuracy of a test? We take a bunch of patients, some of whom have Condition Y, and some of whom don’t, and we run them through Test X like sand through a sieve. Then we see which patients the test flagged, and see how accurate it was. (Obviously, we’ll need a way of knowing for sure who has Condition Y; this is usually done by a separate, “gold standard” test with known reliability. Correlation between Test X and the gold standard is what we’re examining here. Why not just use gold standards tests on all patients? Generally these are difficult, invasive, time-consuming, and expensive procedures — not appropriate for everyone, and certainly not of much use in the field.)

We’ll come up with a couple of figures. One is the test’s sensitivity. This describes how well our test picked up Condition Y; how alert was it, how often did it pick up what we’re looking for? If you have Condition Y, how likely is the test to say you have it? How many sick patients slipped past? If our test has 100% sensitivity, it will have zero false negatives; it will never miss, will never fail to flag a patient with Condition Y. A test with 0% sensitivity is blind; it will never notice Condition Y at all.

The other statistic is the test’s specificity. This describes how selective our test is, how cautiously it sounds its alarm. If you don’t have Condition Y, how likely is the test to say you don’t have it? Will it ever be fooled, and wrongly think that you do? A test with 100% specificity will never produce a false positive; if it shouts positive, it’s never wrong. On the other hand, a test with 0% specificity will never be right; it’s the boy who cried wolf.

Together, sensitivity and specificity describe a test’s accuracy. Intuitively, you can see how the two parameters might often work against each other; we can make a test that is extremely “paranoid,” and will catch almost everything — high sensitivity — but will also flag a great many false positives — low specificity. (Heck, we could just make a flashing red light that said “POSITIVE!” every single time, and we’d never miss anyone — of course, it’d have so many false positives that it’d be useless.) Conversely, we can make a test which is extremely judicious and selective, and when it says “positive,” we can trust that it’s probably right — high specificity — but it’ll miss a lot of true positives — low sensitivity.

Ideally, we’d like a test with high sensitivity and high specificity. But when that’s not possible, then at least we need to understand how to interpret the results.

For instance, a test with high sensitivity is very good for ruling a condition out. Because it almost always catches Condition Y, if the test says “nope, I just don’t see it here,” then that’s very trustworthy; if the patient did have it, the test probably would’ve caught it. Think SnOut: a test with good Sensitivity that comes back negative rules a condition Out.

Example: pinpoint pupils. For the patient with altered mental status, this is a very sensitive indicator of opiate use; almost everyone with a large amount of opiates in their system will present with small pupils. However, it’s not very specific, because many people will have small pupils without using narcotics (for instance, due to bright lighting). So if you don’t see pinpoint pupils, that finding rules out opiate overdose with fairly good reliability.

On the other hand, a test with high specificity is very good for ruling a condition in. Because it’s almost never wrong, if it says you do have Condition Y, you can take that to the bank. Think SpIn: a test with good Specificity that comes back positive rules a condition In. (Thanks to Medscape for these mnemonics.)

Example: a pulsating abdominal mass is an extremely specific finding in abdominal aortic aneurysm. Very few other conditions can cause such a pulsating mass, so if you find one, you can pretty reliably say that the patient has a AAA. However, many AAA patients will not have such a mass, so this is not very sensitive. But if you do find a pulsating mass, this rules AAA in fairly well.

 

Warning: Scary Statistics Ahead

Okay, that wasn’t so bad, was it?

Here’s where things get a little weirder. If you’re barely hanging on to the thread so far, you have permission to stop reading now.

Sensitivity and specificity are the most commonly used parameters describing the accuracy of a test. They’re properties of the test itself, so you can hang those numbers on it and they won’t change on you.

However, anyone who’s studied Bayesian statistics will understand that the true accuracy of our test is not only a factor of the test, but also depends on the prevalence of Condition Y in the population. If Condition Y is exceptionally rare in the patient group we’re looking at, then even if Test X is very specific, it will produce a large number of false positives. Conversely, if Condition Y is exceptionally common, then even if Test X is very sensitive, it will produce a large number of false negatives.

The reasons for all of this are complex. (For some additional reading, see here, and here.) But the general gist is this: if Condition Y is very unlikely to be present (either because it’s generally uncommon, such as scurvy; or because it’s an improbable diagnosis for the individual patient, such as an acute MI in an 8-year-old), then even if your test “rules it in,” it will still be unlikely. The positive test made it more likely, but it was so improbable to begin with, the odds didn’t change very much. And if Condition Y is very probable (such as a healthy heart in an asymptomatic teenager), then even if your test “rules it out,” the odds still support its presence.

What this all means is that in order to answer our real questions, we need another measure. The positive predictive value (PPV) and negative predictive value (NPV) are the answer, and really, these figures are what we’re after. The PPV answers: given a positive test result, how likely is the patient to have the condition? The NPV answers: given a negative test result, how likely is the patient to lack the condition? In other words, in a real patient, how likely is the test result to be correct?

The trouble is that PPV and NPV aren’t just characteristics of the test; as we saw above, they also depend on the prevalence of the condition, or the “pre-test probability.” What this means is that although the study you’re reading may report predictive values, they are not necessarily applicable to your patient. They’re only applicable to the patient population that was studied. Now, if your patient is similar to that population — in other words, has about the same pre-test probability of the condition as they did — then the predictive values should be correct. If not… not so much.

So do we have any more tricks? We have one more: likelihood ratios. Likelihood ratios factor out pre-test probability, producing a simple ratio that describes how much the test changed the probability. For instance, suppose we have a patient who we judge has a 10% probability of having Condition Y. We apply a test with a positive likelihood ratio of 5, and it comes up positive. What’s that mean? The math is a little bit roundabout, because we need to convert probability (a percentage of positive outcome out of all possible outcomes) into odds (a fraction of positive outcome over negative outcome): 10% is the same as 1:9 odds. 1/9 times 5 is 5/9, and if we convert that back to a percentage (positive outcome over total outcomes, or 5/14), we have  the result: about 36%. The patient now has a 36% chance of having Condition Y. Conversely, suppose it came up negative, and the test had a negative likelihood ratio of .1. The post-test probability (by the same calculation) is now only around 1%.

It’s a simple device that would be far more intuitive without the odds vs. probability conversion, but suffice to say that a likelihood ratio of 1 (1:1) changes nothing, higher than 1 is a positive test (1–3 slightly so, around 5–10 is a useful test, and over 10 is highly suggestive), and less than 1 is negative (1–.5 just barely, around .5–.1 decently, and under .1 is strongly negative.) Try plugging numbers into this calculator to experiment — or drag around the sliders in the Diagnostics section at The NNT. The only bad news is that you still need to know the pre-test probability, but the good news is that you can come up with your own estimate, rather than having an inappropriate one already included in the predictive values.

How to come up with pre-test probabilities? Well, research-derived statistics do exist for various patient groups… but realistically, in the field, you will need to wing it. Taking into account the whole clinical picture, including history, physical exam, and complaints, how high-risk would you deem this patient? You don’t need to be exact, but you should be able to come up with a rough idea. Now, apply your test, and consider the results — about how likely is the condition now? If at any point, you have enough certainty (either positive or negative) to make a decision, then do it; there’s no point in tacking on endless tests if they won’t change your treatment.

Anybody still breathing? We’ll talk about odds ratios, NNT, and other intervention-related numbers another time.

[Edit 5/15/13: the follow-up post on outcome metrics is posted at Lit Whisperers, our sister blog]