Archives for January 2012

Understanding Shock X (supplement): Fluid Choices

Although it may not be immediately relevant to most of us prehospital folks, the ongoing battle for supremacy in the world of IV fluids is a fascinating topic that’s worth following. We know that blood is the good stuff, but we remain interested in concocting an artificial fluid that can replace volume and mitigate the shock response — maybe even carry oxygen or support clotting — yet remain logistically feasible for everyday use. The current contenders are:


Normal Saline (aka NS)

Probably the most common fluid used today, this is nothing more than sterile water with .9% NaCL (table salt) dissolved in it. This amount of solute more or less approximates the concentration of our body’s water, which makes normal saline “isotonic”: its tonicity is approximately equal to our cells, making its osmotic pressure very low. In other words, it’s basically the same raw liquid we already have circulating, so its volume of distribution — the amount of saline that will leave the intravascular space, once we drip it in there — is relatively low.

That doesn’t mean we don’t lose a lot, though. Once it’s had a chance to settle out, quite a bit of infused saline will end up in the interstitial space. Typically this distribution will be in the ballpark of 1:3–1:4 — in other words, if we give a liter of saline, within an hour or so only about 250–300ml will remain in the intravascular space. Sicker people (who have problems like increased capillary permeability) have even higher volumes of distribution.

The benefits of normal saline: it’s very cheap. It’s very stable, lasting approximately forever on the shelf, and has minimal storage requirements. It’s compatible with every patient and every med. It’s easy to administer (any access will do, preferably large-bore).

The downsides: it carries no oxygen, impedes clotting, promotes inflammation, produces acidosis (called a hyperchloremic acidosis, since it’s secondary to the chloride content), and generally does absolutely nothing for you except increase the intravascular volume, and it does only an okay job at that.


Lactated Ringer’s (aka Ringer’s Lactate)

This stuff is basically normal saline with some extras. Like NS, it’s isotonic, so the volume of distribution is the same. But in order to mitigate the acidosis produced by NS, it’s got lactate added. Lactate converts to sodium bicarbonate in the blood, and bicarb is a strong base, so Ringer’s essentially comes “buffered” — it should have less impact on the pH. This is good, and large volumes of this stuff have a more benign effect than large volumes of saline. (Ringer’s also includes some other electrolytes, such as potassium and calcium, bringing it closer to the composition of blood serum.)

The downsides: for many prehospital services, the main “downside” is that they don’t want to stock multiple types of fluid, so once they’ve stacked NS on the shelves they’re done. Ringer’s is not as appropriate for general use, since it’s incompatible with some medications and contraindicated in some patients. There is also an old belief that it’s incompatible with blood products — that is, if you hang a bag of PRBCs on your Ringer’s line, the calcium in the Ringer’s will stimulate the coagulation cascade (PRBCs are usually stored by adding citrate, which prevents clotting by binding up calcium) and create emboli. This is now generally understood to be false.


Hypertonic solutions

Now we get into the more interesting stuff.

Remember we agreed that normal saline and Lactated Ringer’s are isotonic? What if we use a fluid that is hypertonic? This would mean that the fluid has a higher tonicity (more dissolved stuff) than our cells. Since the golden rule of osmosis is that water moves toward the space with the higher concentration of dissolved solids, adding hypertonic fluids to the blood — and hence making the blood hypertonic — will cause fluid to move from the intracellular into the intravascular space.

Why would this be good? Well, for one thing, it yields an awesome volume of distribution. Compared to the isotonics, distribution is actually reversed; we end up with more than we put in, not less. Infusing a liter of a typical hypertonic can yield an eventual volume increase of nearly 8 liters.

Isn’t it bad to suck fluid out of our cells? It would seem like it. However, for short-term use (such as emergency trauma care), the effects of this generally seem to be benign. In fact, there is some evidence that using hypertonic solutions may attenuate the inflammatory response associated with fluid administration — perhaps just because we don’t need to give as much of it.

So far, there’s insufficient evidence for the routine use of hypertonic fluids in the civilian world. So far, the research suggests that they’re “at least” as good as the isotonics. The military is another story, though; they love this stuff, because it’s light. Whether or not they should be doing that, in order for a combat medic to dump 4 liters of saline into someone, he’d have to carry 4 liters of liquid on his back — alongside absolutely everything else he’s going to need. Much better to bring some easily-portable 250ml bags of a hypertonic. It’s like an expand-o-fluid.

There are various hypertonics out there, including high-concentration salines (such as 3.0% — call it abnormal saline if you want to be cute) and others. So far nothing’s really landed on top, although mannitol is often used to suck fluid from the brain and cause “shrinkage” during cerebral edema.



Saline is a crystalloid fluid because it’s water with small ions dissolved in it. The sodium (Na) and the chloride (Cl) are not like particles of sand, swirling around in there but too small to see — they’re fully dissolved and dissociated.

Colloids are different. A colloid is a large molecule, something too big to easily cross cellular membranes. These don’t dissolve in the same way; they’re more like ice cubes rattling around in your glass. Blood itself is a colloid, since it contains big molecules like red blood cells.

“If blood is colloidal,” the wags say, “why not try giving colloidal fluids?” Well, all right then.

One big benefit of this would be the volume of distribution. Since the colloidal solids can’t easily escape across the membranes, they remain in the intravascular space and hence keep the oncotic pressure high.

But they’re usually expensive. And tend to be more complicated (in indications and contraindications) than crystalloids. And can be more finicky to store. And for the most part, have been shown to be no better than crystalloids. Oh well.


Artificial oxygen-carrying colloids

Well, here’s a neat idea. Maybe an arbitrary colloid isn’t much good, but can we make one that mimics blood — can we come up with a fluid that actually binds and carries oxygen in the same sort of way as our red blood cells? If we could create such a thing, and if it were broadly compatible and not too expensive and had a reasonable shelf-life, it would be the next best thing to using blood and a major breakthrough.

We have created such things, either wholly artificial or derived from purified (usually cadaverous) blood samples. You can store them for ages, although they’re not particularly cheap, being new, on-patent drugs. So far they all seem to have little to no benefit in outcome — and often an increased rate of complications like heart attacks. Hmm. The search continues. (The trick may be to come up with something that shares more of blood’s qualities, such as positive-feedback binding, and maybe even some clotting goodness. We’ll see.)


Hypotonic fluids 

Like half-normal saline! Good stuff, right? Wait, no. That would have a god-awful volume of distribution. Excellent, you’re paying attention.


Blood Products

You really were paying attention! Full circle we come. Although blood is not all things to everybody, and has its own negatives and caveats, at the present date if you lose blood the best replacement is blood. Of some kind.

Of what kind remains a bit of a mystery. Men in white coats continue to play with different mixtures of red cells, and plasma, and platelets, and even various concentrates and precipitates of specific clotting factors. One of the latest miracle additions is tranexamic acid, which antagonizes natural thrombolytics (remember plasmin?) and seems to reduce bleeding. There are also cool devices, used mainly during surgery, that “salvage” your own lost blood, rinse it off, and give it right back to you, which obviously simplifies some things.

Of note is an approach to transfusion developed by the anaesthesiologists at Shock Trauma in Baltimore. They like to give PRBCs and plasma until you reach a reasonably permissive pressure. Then they bolus some opiate goodness (fentanyl is nicely controllable). This puts a brake in the patient’s compensatory catecholamine response — their clamped-down veins and arteries relax a little. Which drops the pressure again. So they give some more fluid. Which raises the pressure again. Then they give more fentanyl. Repeat repeat repeat. The end result? A well-resuscitated patient — with a nice pressure — but with a relaxed, normal vasculature — and a normal volume. It’s not hard to fill up a severely compensating patient; their pipes are tiny. But it’s also not as good as filling them up to a normal perfusing volume. Neat idea. (Plus, pain management or sedation for surgery is no problem with that much fentanyl on board!)

Best of all, of course, is simply not to lose the blood to begin with. Tourniquets have really made a resurgence, and many feel that at this date, nobody with reasonably timely medical care should ever die from an extremity injury — not if you can slap a tourniquet somewhere proximal and cinch it down until the bleeding stops. The military has led the way with this, as with the use of hemostatic agents — powders you sprinkle on (or, nowadays, often come pre-embedded in a dressing) which help chemically promote clotting when combined with direct pressure.


Okay, so where does all of this leave us?

We’re not sure. Despite decades of research into this topic, best practices remain uncertain. But the following are probably true:

  1. Extremes are probably to be avoided. Too much or too little of anything is rarely good.
  2. If there is any benefit for non-oxygen-bearing, non-clotting fluids in hemorrhagic resuscitation, it is likely limited to a supplemental or temporizing role.
  3. Further evidence may or may not demonstrate a benefit from hypertonic solutions.
  4. A really usable “instead of blood” fluid remains the holy grail, and is not yet available.

and most of all…

  1. There are significant negatives associated with any fluid administration, so in order to produce real improvements in survival, any benefit must be substantial enough to outweigh this basic harm.

Thanks to everyone who stayed with us through this lengthy chat about shock! I want to give particular thanks to Dr. Jeffrey Guy, whose teachings were instrumental in forming the core of my own material.


Back to Part IX

Understanding Shock IX: Assessment and Recognition

To wrap up our story on shock, let’s discuss how to recognize it.

We all have some idea what shock looks like. Like many pathologies, its loudest early markers are actually indirect — we’ll often recognize the body’s reactions to shock rather than the shock itself.

Although there are a few ways to classify the stages of shock, let’s just use three categories here.


Early or Insignificant

Shock that is very early or minimal in effect may have no particular manifestations. One situation where significant or late shock may also be “hidden” is in the elderly patient, or anyone with significant comorbidities; if their body’s ability to mobilize its compensatory mechanisms is poor, then the red flags won’t be as obvious. This doesn’t mean the shock isn’t as bad; in fact, it means that it’s worse, because their body can’t do as much to mitigate it.

The way to recognize shock at this stage is from the history. If we see an obvious bullet hole in the patient’s chest, and three liters of blood pooling on the ground beside him, then it doesn’t matter how the patient presents otherwise; we’re going to assume that shock is a concern. Blood volume is proportional to bodyweight, but for a typical adult, a fair rule of thumb is to assume about 5-7 liters of total volume. (Not sure what a liter looks like? The bags of saline the medics usually carry are a liter; so are those Nalgene water bottles many people drink from. “Party size” soda bottles are two liters.) Losing more than a liter or two rapidly is difficult to compensate for.

Remember, of course, that blood can also be lost internally, and aside from the occasional pelvic fracture or hemothorax, the best environment for this is the abdomen. Always examine and palpate the abdomen of the trauma patient, looking for rigidity, tenderness, or distention. Remember also that the GI tract is a great place to lose blood; be sure to ask your medical patients about blood or “coffee grounds” (old blood) in the vomit or stool.

Fluid enters and leaves the body continuously, and any disruption in this should be recognized. If a patient complains “I haven’t been able to eat or drink anything in two days,” they’re telling you that they haven’t taken in any fluid for 48 hours. If they tell you they’ve been vomiting or experiencing profuse diarrhea, that’s fluid leaving their body in significant volumes. What about the man who just ran a marathon and sweated out a gallon? Did he drink a gallon to replace it?


Compensated Shock

Significant shock will result in the body attempting to compensate for the low blood volume. Much of this work is done by the sympathetic system, and there are two primary effects: vasoconstriction and cardiac stimulation.

By constricting the blood vessels, we can maintain a reasonable blood pressure and adequate flow even with a smaller circulating volume. We normally vasoconstrict in the periphery — particularly the outer extremities and skin — “stealing” blood from those less-important tissues and retaining it in the vital core. This causes pallor (paleness) and coolness of the external skin. The sympathetic stimulation may also cause diaphoresis (sweating), which is not compensatory, but simply a side effect of the adrenergic release.

The heart also kicks into overdrive, trying to keep the remaining volume moving faster to make up for the loss. It beats faster (chronotropy) and harder (inotropy), resulting in tachycardia. Note that patients who use beta blockers (such as metoprolol) may not be able to muster much, if any, compensatory tachycardia.

A narrowing pulse pressure (the difference between the systolic and diastolic numbers) may be noted; since the diastolic reflects baseline pressure and the systolic reflects the added pressure created by the pumping of the heart, a narrow pulse pressure suggests that cardiac output is diminishing (due to loss of preload), and that more and more of the pressure we’re seeing is simply produced by shrinking the vasculature.

Tachypnea (rapid respirations) are also typically seen. In some cases, this may be due to emotional excitement, and there is also a longstanding belief that it reflects the body’s attempts to “blow off” carbon dioxide and reduce the acidosis created by anaerobic metabolism. (Interestingly, lactate — a byproduct of anaerobic metabolism — can be measured by lab tests, and is also a sign of shock, particularly useful in sepsis.) Additionally, it ensures that all remaining blood has the greatest possible oxygenation. However, it is also plausible that this tachypnea serves to assist the circulatory system: by creating negative pressure in the thorax (the “suction” you make in your chest whenever you inhale) and positive pressure in the abdomen (due to the diaphragm dropping down), you “milk” the vena cava upward during inspiration, improving venous return to the heart and allowing greater cardiac output. This “bellows” effect helps the heart fill more and expel more with each beat.

The more functional the patient’s body is — such as the young, strong, healthy victim — the more effective these compensatory systems will be. Hence the old truism that pediatric patients “fall off a cliff” — they may look great even up through quite profound levels of shock, due to their excellent ability to compensate, then when they finally run out of room they’re already so far in the hole that they become rapidly unhinged. It’s great that these people can compensate well, but it does mean we need to have a high index of suspicion, looking closely for signs of compensation (such as tachycardia) rather than outright signs of shock — because by the time the latter appears, it may be very late indeed.

Patients in compensated shock may become orthostatic; their bodies are capable of perfusing well in more horizontal postures, but when gravity pulls their remaining blood away from the core, this added challenge makes the hypovolemia noticeable. Less acute shock due to causes like dehydration may result in dry skin (particularly the mucus membranes; try examining the inside of the lower eyelid) with poor turgor (pinch a “tent” out of their skin and release it; does it snap back quickly or sluggishly?), and potentially with complaints of thirst. Urine output will usually be minimal. Generally, the more gradually the hypovolemia sets in, the more gradually it can be safely corrected; it’s the sudden, acute losses from causes like bleeding that we’re most worried about.


Decompensated Shock

As shock continues, compensatory systems will struggle harder and harder to maintain perfusion and pressure. Eventually they will fail; further vasoconstriction will reduce rather than improve organ perfusion, beating the heart faster will expel less rather than more blood, and the blood pressure will start to drop.

The hallmark of this stage of shock is the normal functioning of the body beginning to fail. The measured blood pressure will decrease and eventually become unobtainable. Pulses will weaken until they cannot be palpated. As perfusion to the brain decreases, the patient’s mental status will deteriorate. Heart rate and respirations, previously rapid, will begin to slow as the body loses the ability to drive them; like a government office that can’t pay its workers, the regulatory systems that should be fighting the problem begin to shutter their own operations. As the heart continues to “brady down,” eventually it may lose coherence (ventricular fibrillation), or keep stoically trying to contract until the last, but lose all effective output due to the lack of available blood (PEA). Cardiac arrest ensues, with dismal chances for resuscitation.


Alternative Forms of Shock

Although we have focused so far on hypovolemic shock, particularly of traumatic etiology, there are other possibilities. A wide range of shock types exist, but speaking broadly, there are only two other categories important to us: distributive, and cardiogenic/obstructive.

Distributive shocks include anaphylactic, septic, and neurogenic. The essential difference here is that rather than any loss of fluid, the vasculature has simply expanded. Rather than squeezing down on the blood volume to maintain an appropriate pressure, the veins and arteries have gone “slack,” and control of the circulating volume has been lost; it’s simply puddled, like standing water in a sewer pipe. (Depending on the type of shock there may also be some true fluid losses due to edema and third-spacing.) Imagine tying your shoes: in order to stay securely on your feet, the laces need to be pulled snugly (not too tight, not too loose). If the knot comes undone and the laces lose their tension, the shoe will likely slip right off. Your foot hasn’t gotten smaller, but the shoe needs to be hugging it properly to stay in place, and it’s no longer doing its job.

The hallmark of distributive shock is hyperemic (flush or highly perfused) rather than constricted peripheral circulation. The visible skin is warm (or hot) and pink (or red), and the patient may be profoundly orthostatic. Septic shock is associated with infection; anaphylactic with an allergic trigger; and neurogenic with an injury to the spinal cord.

Cardiogenic and obstructive shocks are a different story. In this case, there’s nothing wrong with the circulating volume, or with the vasculature it flows within; instead, there’s a problem with the pump. Cardiogenic shock typically refers to situations like a post-MI heart that’s no longer pumping effectively. Obstructive shock refers to the special cases of pericardial tamponade, massive pulmonary embolism, or tension pneumothorax: physical forces are preventing the heart from expanding or blood from entering it, and hence (despite an otherwise functional myocardium) it’s unable to pump anything out. In either case, we can expect a clinical picture generally similar to hypovolemic shock, but likely with cardiac irregularities — such as ischemic changes or loss of QRS amplitude on the ECG, irregularity or slowing of the pulse, or changes in heart tone (such as muffling) upon auscultation. Pulsus paradoxus (a drop in blood pressure — usually detected by the strength of the palpable pulses — during the inspiratory phase of breathing), electrical alternans (alternating QRS amplitudes on the ECG), and jugular vein distention also may be present in the case of tamponade or severe tension pneumothorax.


In sum, remember these general points:

  1. The history and clinical context should be enough to make you suspect shock even without other signs or symptoms.
  2. The faster the onset, the more urgent the situation; acute shock needs acute care.
  3. Look both for signs of compensation (such as tachycardia) and for signs of decompensation (such as falling blood pressure). However, remember that due to confounding factors (such as particularly effective or ineffective compensatory ability, or pharmacological beta blockade), any or all of these may be absent.
  4. Distributive shocks are mainly characterized by well-perfused peripheral skin; cardiogenic/obstructive shocks are characterized by cardiac irregularities.

Interested parties can stay tuned for a brief appendix discussing fluid choices for resuscitation — otherwise, this journey through shock is finally finished!


Go to Part X (appendix) or back to Part VIII

Now with Organization!

Although somewhat late out of the gate, we’ve finally put together an organized, all-inclusive site index for archived posts. Browse by topic and brief description to find what you’re looking for; we’ll make an effort to update the index regularly with new material, and it’ll be periodically reorganized as needed. Have at it!

Understanding Shock VIII: Prehospital Course of Care

Now that we have a pretty good idea of how shock works, what does it all mean for our treatment in the field?

Much like cardiac arrest and some of the other “big sick” emergencies, there are really a couple essential interventions we need to execute, maybe a couple others that aren’t a bad idea, and beyond that, our main job is to ensure that we don’t kill our patient by wasting time doing anything else.


Step 1: Control the bleeding

As we emphasized ad nauseam, the number one goal with the bleeding patient is to stop the bleeding. No need to beat this to death, but just remember: if you can control the bleeding, yet don’t get much of anything else done, you’re doing absolutely fine.


Step 2: Transport to surgery

In most significant cases of hemorrhage, definitively controlling the bleeding will require surgical intervention. We don’t do surgery, but we do set the stage, which is why it’s essential for us to know what we’re doing. Get thee to a trauma center, and quickly!

Can other hospitals perform surgical intervention? Sometimes. Maybe. A world-class trauma surgeon might happen to be in the building for a conference. Maybe the operating room is between scheduled procedures and happens to be clean and available. But the point to a trauma center is that it’s guaranteed to have certain resources available, and that’s the kind of place we want to bring these patients. 9 times out of 10, if we transport them elsewhere, they’ll simply end up being transferred back out to the trauma center anyway, making the whole exercise essentially one very long transport. Can a small community hospital help stabilize the patient before surgery? Sure — but as we know, everything else is a distant second priority to bleeding control. Even transfusing blood may need to be done sparingly until the leak has been corked.

What about ALS? Do these patients need paramedics? Now, if they acutely decompensate and need airway management or other interventions you can’t provide (or have other issues like pneumothorax), then ALS-level care would be valuable. But outside of that, and even granting that to a certain extent, a medic unit is not going to stitch up the bleeding, and meeting them will certainly delay transport to surgery at least by a few minutes. True, they’ll be able to initiate IV access that can be used for blood later, but in most cases this takes mere seconds at the ED (where there’s plenty of room, good lighting, and ample personnel) — and prehospital IVs will sometimes be replaced anyway.


Step 3: Promote oxygen delivery

Okay, you shock technician, now what?

Can we talk about coagulopathy of trauma — aka the “deadly triad”?

Bleeding control is the priority, right? And bleeding control requires clotting. But there’s a set of conditions guaranteed to obstruct clotting, and three of them are almost always present during hemorrhagic shock.

One is hemodilution. When we top off our bleeding patients with non-blood fluids, as we’re so fond of doing, it dilutes both oxygen-carrying capacity (since we’re not adding red blood cells) and clotting speed (since we’re not adding platelets or clotting factors). So this one’s our fault, and can be readily avoided by simply resisting the urge to replace blood with salty water.

One is acidosis. If you’ve been paying attention, you know that acidosis tends to develop in shock due to anaerobic cellular activity, and can be further encouraged by overzealous fluid administration. Is this the end of the world? (After all, a little acidosis might even improve oxygen delivery by shifting the oxyhemoglobin dissociation curve.) Well, the trouble is that acidosis also leads to coagulopathy. According to some in vitro studies, in fact, even mild acidosis can precipitously decrease platelet aggregation, and in significant acidosis platelets won’t activate at all. Zero.

The last is hypothermia. Not only do cold patients have poor oxygen delivery and other problems, they clot poorly; low temperatures cause coagulopathy too.

Now, we can’t do much about the initial trauma. We can discourage acidosis by limiting fluid use, and ensuring that ventilations remain adequate. What about hypothermia? Do our trauma patients get cold? What would you expect when you take someone who’s bleeding, strip them naked on a cold sidewalk, pump cold saline into their veins, and chuck them into an ambulance carefully heated to your comfort?

Keep your trauma patients warm. This is not about human kindness or TLC, this is a serious and important intervention for shock. Hypothermia is great for cardiac arrest, it may be beneficial in some other scenarios, but it is not good for bleeding people.

How about supplemental oxygen? Well, I suppose so. In the patient with adequate respirations, it is doubtful that “topping off” their PaO2 will affect them appreciably; but as they begin to decompensate, they’ll need all the help they can get.

Positioning? Remember how big a deal they made about the Trendelenburg position in school — how it pulls blood from the lower extremities into the core? And ever noticed how it’s not exactly our number one emphasis in the field? Trendelenburg has little real evidence supporting it, and the bulk of what does exist suggests its effect is fairly minimal — it moves only a little blood, the effect is transient, and the body’s compensation can actually cause a paradoxical reduction in core perfusion. Mostly these studies were done in healthy people, so it’s possible that our shocky patients do get a little benefit — and one supposes that if things are dire enough to need every last cc of blood, you can give it a shot. But typically it won’t do you too many favors. (I certainly wouldn’t advise propping the patient bolt upright, though!)


Step 4: Supportive care

Supportive care means battling secondary problems as they arise.  It doesn’t mean waffling over nonsense while your patient bleeds out.

If the patient’s airway is compromised, or you have legitimate reason to think that it may become compromised, then it should be managed. If they’re breathing inadequately, they’ll need assistance. Beyond that, any other care should only occur after you’ve stuck a cork in the bleeding and started rolling toward the guys with knives. Cardiac fiddling, pain management, splinting or minor bandaging — these should take place en route or simultaneous to other care, if at all. Shock kills people; is a nice sling-and-swath going to save them?

Spinal immobilization? It’s been pretty definitively shown to hurt rather than help in penetrating trauma. What about combined blunt and penetrating? There’s no evidence that it helps and some evidence that it’s harmful. We have no reason to think that tying people to boards does anything good, but we do know that wasting time here does everything bad. So if your local protocols demand immobilizing these patients, I won’t tell you otherwise — but please, at least, try and hurry.

That’s it, folks. Let’s wrap it all up next time by talking about recognizing the beast.

Key points:

  1. Stop the bleeding to the greatest extent possible in the field.
  2. Immediately and without delay transport to a facility capable of emergency surgery.
  3. Provide other supportive care as necessary, without delaying #1 and #2.
  4. Maximize oxygen delivery with supplemental O2, keeping the patient warm, and consider the Trendelenburg position.
  5. Minimize delays created by any and all non-essential care.


Go to Part IX or back to Part VII

Understanding Shock VII: Negatives of Fluid Resuscitation

The last time we talked, we learned about the arguments in favor of non-blood fluid resuscitation. What are the arguments against it?


The “blow out the clots” argument

The vascular system is a pressurized circuit. Bleeding means poking an opening in this circuit, and we know that repairing this hole is our number one priority.

The body is pretty good at fixing leaks in its vasculature. But it’s not magic. It’s going to try to form a stable clot that covers and seals the hole, just like wrapping tape around a leaky pipe fitting.

What’s a good way to make this task harder? Increase the pressure inside the pipe. The faster that blood wants to rush out of the hole, the tougher it’s going to be to get a clot to stick there.

Imagine your inflatable raft has a pinhole in it, so you cover it with a piece of tape. It seals well. Then you drop a cooler of beer onto the raft, increasing the internal pressure. The tape blows off. Simple.

Many providers have therefore moved towards the practice of permissive hypotension — resuscitating only to a lower than normal blood pressure — and/or delayed resuscitation — waiting for substantial fluid replacement until bleeding has been controlled. Permissive may mean a pressure of 80, 90, or 100; it may mean giving crystalloids sparingly and only until blood becomes available; or it may mean giving nothing at all except the good stuff. Or you can take a page from the military, which says to resuscitate until a radial pulse is palpable, and the patient’s mental status is restored — then stop.


The dilution argument

There’s another reason why filling the patient with salt water might make it harder to control their bleeding.

Their body is trying to build clots at the location of injury. We want to encourage this process. In order to occur, it requires the activity of circulating platelets and clotting factors.

Mixing the patient’s blood with saline increases its volume but doesn’t increase the number of these clotting precursors. In other words, we’re diluting their blood, just like a bartender watering down your drink. There’s more volume in your cup, but there’s no more of the stuff we care about. And since the ability to form clots is closely related to the concentration of the clotting components, diluting the blood means slower clotting.

Together, these two arguments form a compelling case against the “volume for the sake of volume” theory. The patient’s ability to form clots and stop the bleeding isn’t a small thing; in a way, it’s the only thing. In fact, INR (a measure of clotting speed) has been shown to be a key predictor of whether a trauma patient will survive their injuries.


The proinflammatory argument

One of the key forces in the shock cascade is inflammation. So it seems like promoting more inflammation is the last thing we’d want.

But surprise: infusing fluids can do exactly this. It’s not entirely clear why this happens, but it’s unquestionably true; fluids encourage the inappropriate immune response and increase inflammation and tissue dysfunction. Suffice to say that this is bad.

Back in Vietnam, when aggressive fluid resuscitation really became trendy, doctors were perplexed to find many of their volume-resuscitated patients with a severe condition called “Da Nang lung” (nowadays Acute Respiratory Distress Syndrome) — wet, failing, edematous lungs with no cardiac cause. The combination of increased fluid volume plus increased inflammation means failing lungs. Or check your nearest ICU to see some abdominal compartment syndrome, where fluid fills the abdomen until the organs fail. What were you were saying about fluids being harmless?


The acidosis argument

The pH of our bodies is a hair over 7. Pick up the nearest bag of normal saline and read the label. What’s its pH?

Is it 7? No? More like between 5.0 and 6.0? Interesting. Remember that pH is a logarithmic scale, so we’re talking a difference of 10–100 here. So that nice “normal” fluid can promote significant acidosis.

Is this bad? Only if you like clotting. Acidosis is detrimental to coagulation (among other things), for reasons we’ll get into later. Clotting is good!


The what’s-the-point? argument

In the end, the most compelling argument against pouring what amounts to water into trauma patients is this: fundamentally it is not what they need. Their problem is not a lack of normal saline. “When I find a patient who’s bleeding crystalloid,” some providers are fond of saying, “I’ll give them crystalloid. But usually, the puddle on the ground is blood.”

Now, in some patients, crystalloid may indeed be what’s missing; we’ll touch upon situations like sepsis and dehydration later. But if they’re bleeding, it seems like — at best — playing with any fluid except those that can restore oxygen-carrying capacity or promote clotting is a waste of time that could be spent patching the hole and rushing toward surgery. And at worst, it may be exacerbating the problem.

For a long time, paramedics were taught to fill the hypotensive patient with fluid until their blood pressure was normal. The jury is still out on the best practices for fluid resuscitation, but there is fairly widespread agreement now that this is a bad idea. Many progressive systems have gone the route of giving no crystalloid whatsoever for hemorrhagic shock, or at least giving it very sparingly. Seeing the numbers 120/80 on the monitor seems like a good thing, but shock is not a blood pressure, raising the blood pressure is not necessarily beneficial, and we’re supposed to be making the patient feel better, not ourselves.

So, stop the bleeding, and restore the stuff that matters. Since we rarely give blood in the field, the first one is the main business of EMS. And oddly enough, it’s very much a BLS skill.


  1. Increasing the blood pressure interferes with bleeding control.
  2. Diluting the blood discourages clotting while doing nothing for oxygen transport.
  3. Aggressive fluid resuscitation promotes inflammation, edema, and organ dysfunction.
  4. Current best practices are unclear, but likely involve a minor role for crystalloid resuscitation, in favor of bleeding control, blood products, and early surgical intervention.

Next time: mastering the field treatment of hemorrhagic shock.


Go to Part VIII or back to Part VI