Skip to content

The Real Immune System

September 9, 2010

The integument is generally underappreciated as the true barrier against infection.

When physicians and medical researchers refer to the immune system, they usually mean the extremely complex system of white blood cells and antibodies that attack and destroy microbial invaders that get into the body’s tissues.  Yet this activation only occurs after there has been a breach, making that system more of a border patrol than one which confers true immunity.

The paradigm shift in medicine is to consider the integument as the true immune system.  The human is wrapped by a continuous sheet of integument: the epithelial lining known as skin is continuous with the epithelium lining the inner lumen of the intestinal tract, the two epithelia joining each other at the mouth and the anus.  These epithelial layers consist of continuous sheets of cells bound to each other by what are term tight junctions, and they sit on a solid basement membrane.

The body is roughly 60% water, which means that we are mostly water.  Yet, we don’t leak.  The reason we don’t leak is that our bodies are wrapped by this continuous sheet of integument.  If water can’t get out, then bacteria cannot get in, because molecules of water are infinitely smaller than bacteria, which themselves are made up of water.  (The sweat and other fluids that flow off the intestine do not represent leaks; they are actively secreted by the lining cells.)

Where we really live is in the interstitium, which houses the cells (nerve, bone, and muscle) responsible for our movement through the environment.  The so-called immune system (i.e., border patrol) also lives in the interstitium seeking defects and invasion.  Unless the integument is violated, there is no work for them to do.

If the integument is intact, we are truly immune.  It takes a breach of the integument, even microscopic, to enable microbes to infect us, thereby activating the so-called immune system.

Reconciling Antimicrobial Choices with Patient Safety Concerns

September 9, 2010

In this era of patient safety, it makes little sense to use antimicrobials with dose-related toxicity unless no alternatives exist.

 Antimicrobials are designed to kill bacteria, fungi, and parasites, which are living organisms.  Human beings, also living organisms, share many metabolic processes with bacteria, so simple poisons can kill humans as well as infecting organisms.  Good antimicrobial design selectively kills the infecting organisms while leaving humans unharmed.

 There have been a number of antimicrobials with toxic potential for humans: chloramphenicol,  B, vancomycin, and the aminoglycosides are key agents that come to mind.  As antimicrobial development has progressed, many of these agents have been replaced with drugs that are significantly less toxic. 

 However, it appears that the replacement process is relatively slow in clinical practice.  For example, relatively safe broad-spectrum penicillins and cephalosporins that could cover Gram-negative organisms emerged in the late 1970s and early 1980s, but did not seriously supplant aminoglycosides as drugs of first choice until the 1990s, when they started to come off patent.  In fact, many orthopedists still employ aminoglycosides as prophylactic agents despite evidence that they are no more effective than more modern drugs.  Similarly, fluconazole was approved by the FDA in 1990 but was heavily restricted by U.S. hospitals due to its high cost, making the highly-toxic amphotericin B the preferred agent.  Under Medicare’s prospective payment system and the managed care practices of private insurers, hospitals could no longer pass on the increased costs to those third-party payers.  It wasn’t until fluconazole came off patent in 2005 that it broadly replaced amphotericin B as the drug of choice for initial antifungal therapy.

 It appears that, despite better safety profiles, initial proprietary drug costs restrict their use under Medicare and managed care systems.  This process is bolstered by clinical practices that favor the perpetuation of traditional management processes and look skeptically at new developments.  The paradigm shift that should occur in medicine is that patient safety should be worth the price we pay.

What the Medicolegal System Needs To Do

September 7, 2010

For decades, the medicolegal system has parasitically fed off the sympathies of juries for injured patients and attacked physicians and hospitals for perceived errors.  In many cases, no errors were committed, but the expectations of patients, families, and the lay public have been unrealistic, expecting real-life outcomes to match the results dreamed up on television dramas.  While there are certainly some poor performers in our industry, the financial penalties in the form of high insurance premiums and the practice of defensive medicine seriously limit the ability to provide improvements to the system.

Without medicolegal reform, quality improvement will continue to be stifled.  Unless, of course, one subscribes to the principle that “the beatings will continue until morale improves.”

There are two fundamental changes that should occur for medicolegal justice to be served:

  • We must demand to be tried in front of a jury of our peers.  It is usually argued that in the United States, we are all peers of one another.  However, this only refers to the socioeconomic class of individuals.  Because we are presumably classless in this country, then everyone is a peer of everyone else.  Thus, an illiterate uneducated homeless individual is the peer of a nuclear scientist on trial for some complex mathematical calculations that led to a toxic leak at a power facility.  It is clear that the socioeconomic concept of a peer is antiquated and ineffective at delivering justice.  Our peers are other healthcare practitioners who better understand the issues at hand and realistic expectations of healthcare.  Human biology is complicated enough that we cannot expect the uneducated to understand the concept sufficiently to make an informed judgement.  Would it make sense for anyone without any mathematics education to determine if a question on a calculus quiz was fair? 
  • We must also demand a Loser Pays system, which exists in nearly every other civilized country.  Through this mechanism, we would see a reduction in a number of lawsuits that appeal to plaintiff’s attorneys because of a sympathetic plaintiff with an adverse outcome from a complicated medical condition (likely to be misunderstood by a lay jury).

These paradigm shifts by the legal system would vastly improve the qualiaty and reduce the costs of health care.  However, they are usually resisted by U.S. trial attorneys, claiming that it would prevent justice for the lower rungs of society (who are usually already getting their medical care paid for by all of us).  Recently, a plaintiff”s attorney who specializes in medicolegal cases for plaintiffs, told me that he lies about his legal specialty when he is seen for his health care.  I recommend that we make these individuals be honest about who they are (so they know we are treating them fairly and without prejudice) by requiring them to sign a petition favoring these legal remedies when they are seen for their health care problems.  We could easily develop a list of who they are from the data available at


September 1, 2010

A paper in the June 2010 Journal of Trauma looks at the accuracy of survival and non-survival predictions on critically ill trauma patients by healthcare providers, with 326 predictions made on 223 patients.  They also looked at the accuracy of various statistical scoring tools (SOFA, APACHE II, RTS, ISS, TRISS).  Overall, health care providers were better outcome predictors than were the scoring systems.  However, nothing provided 100% accuracy.  In fact, of patients that attending physicians predicted would die, 38% actually lived and 5% went home directly without going to a skilled nursing facility or a rehabilitation unit.

For years, I’ve been explaining the possible worst-case scenario to families of severely ill or injured patients.  In our out-of-control, suit-happy society, we no longer give them reassurance that “everything will be all right,” because anything other than a perfect outcome could be grounds for a lawsuit.  By “hanging crepe,” families are prepared for the worst.  They can understand the odds stacked against our efforts to pull their loved one through their brush with death.  In fact, if the patient survives, we look like heroes.

Yet now it seems that I’m seeing a greater number of families throw in the towel long before we’ve given up hope.  Recently, a spouse chose to discontinue life support on a head-injured motorcyclist who’d improved from a GCS of 3T (worst) to 7T (halfway to normal) and was close to extubation.   She made her decision because she feared a chronic vegetative existence for her husband.  I suspect our bleak presentations and television dramas influence them, and that’s unfortunate.  After all, as the study shows, we can’t predict.  After all these years and all our developments, we still can’t tell you what’s going to happen, whether they’ll live or die.  We don’t have a crystal ball.

Our best plan is to proceed with aggression as long as we’re seeing improvements in the patient’s condition.  We should only concede futility when nothing seems to be working.

The paradigm shift in medicine is for families to understand that plan.

Blood Pressure is a Side Effect

August 26, 2010

Much of our clinical practice regarding the circulation involves evaluating and treating a patient’s blood pressure, usually measured in millimeters of mercury.  Blood pressure is the side-wall pressure measured in large arteries in the circulation, usually in vessels in the arms or legs.   We continue to use blood pressure as a stand-in assessment of cardiac output, which is the volume of blood flowing through the circulation over a period of time, typically in liters per minute. 

The pressure inside a pipeline is directly proportional to the amount of flow going through it: the higher the flow, the higher the pressure.  However, a pipeline has rigid pipes.  Human “pipes”, our arteries, are flexible.  They can stretch and contract, making the lumen larger or smaller, respectively.  So if our vessels change their size and the flow rate is no different, we will observe a lower blood pressure with a larger (vasodilated) set of vessels and a higher blood vessel with a smaller (vasoconstricted) set of vessels.  Thus, the use of blood pressure to reflect flow is flawed because of the inherent assumption that the blood vessels are rigid.

The problem is that those assumptions are not considered in standard clinical practice.  At best, a low blood pressure should be seen as a screening test.  When the blood pressure drops, an investigation should be undertaken to determine which of the various forms of shock (hypovolemic, cardiogenic, septic) is at work so that the problem can be fixed appropriately.  It makes little sense to provide increased afterload to someone who is in cardiogenic shock or to vasoconstrict the already vasoconstricted vessels that develop spontaneously in hypovolemic shock.  Yet, because we’re focused on fixing the low blood pressure, that is exactly what we do.  In fact, a recent and noteworthy clinical study continue to treat them as one.

A high blood pressure actually indicates that the heart is working too hard and the vasculature can become damaged.  Our current management is reasonable, as we seek to reduce global blood pressure and the risk of hypertensive heart disease.  However, there is a risk that a widespread reduction in blood pressure could impair the flow to some vital organs.  The paradigm shift in medicine will be to identify the abnormal mechanisms that drive the heart to pump so vigorously and treat those specifically.

Daily Labs Are Obsolete

August 18, 2010

Point-of-care testing should change our routines for patient assessment.

 Traditionally, physicians have ordered daily laboratory tests, especially for their most critically ill patients.  In most hospitals, these blood counts and electrolyte levels are drawn early in the morning, often at 4 am or earlier, waking patients from their sleep.  In contrast to newer technologies, this practice has never been seriously challenged despite the lack of data demonstrating that routine daily labs improve outcomes or reduce costs.  It’s not even certain the tests pay for themselves.

 Routine daily laboratory testing is simply a tradition from the days when it took a few hours to have their results returned.  Physicians making morning rounds felt they needed to see the results at the time of rounds in order to make a full assessment of the patient’s status.  This could only be accomplished by having the laboratory tests drawn early in the morning, often dispatching teams of phlebotomists throughout the hospital, and batch running all these tests at once in an increasingly overburdened laboratory.  In truth, many of the tests run may not have been absolutely necessary, but were done only to be sure that everything was available when the physician assessed the patient on rounds.

 A paradigm shift in medicine is long overdue.  Point-of-care testing (POCT) has been available for most of the routine laboratory tests performed for over a decade.  POCT is accurate and rapid, typically only taking a few minutes to generate the results.  Yet few hospitals and physician practices have completely latched onto these techniques.

 The real improvements in quality will come when POCT actually changes practice.  Selective testing based upon the clinical examination of the patient should be the approach.  Because turnaround is so rapid, no routine daily lab testing would be necessary.  Instead, the physician should test at the time of rounds using POCT.


August 11, 2010

Understanding the fundamental nature of shock will be a paradigm shift in medicine that will enable more effective resuscitation of our sickest patients.

When a surgeon, intensivist, or other acute care practitioner uses the term “shock”, they are describing something that embodies inadequate energy supply to the body’s living cells.  There have been various descriptions and classifications of shock over the past several decades.  One of the most enduring classifications was by Alfred Blalock in 1937, who defined four principal categories of shock: hypovolemic, cardiogenic, neurogenic, and vasogenic.  To a large degree, this classification has persisted for over 70 years, with some modifications (for example, vasogenic became septic) and additions (like obstructive, anaphylactic, and distributive).   Classification has been helpful because it can guide appropriate treatment.  For example, we give volume for hypovolemic shock because that’s what’s needed.  However, volume administration can exacerbate cardiogenic shock; diuretic drugs to eliminate excess volume and other measures that help the heart pump better are best.

In the case of hypovolemic and cardiogenic shock, the reduced cardiac output delivers inadequate oxygen and nutrients relative to the net tissue demand for those fuels.  In septic shock, the circulation is actually working at a higher level than normal; the fact that we can see evidence of poor cellular metabolism in sepsis therefore suggests that cells are somehow poisoned and unable to use the fuels being delivered. 

Neurogenic shock, on the other hand, is not really shock at all, because energy supply to tissues has not changed.  With pure neurogenic “shock”, patients do not develop the signs of tissue hypoxia, such as acidosis or lactate accumulation.  They simply manifest hypotension, which, unfortunately, the modern-day clinician still believes is causative of tissue hypoperfusion.

Managing Shock

August 4, 2010

In order to correct a problem effectively, you need to understand the nature of the problem and what went wrong to produce it. 

Problems often become evident when something happens to get our attention.  But correction of the problem usually involves doing something about the underlying cause rather than the signal we received.  For example, if you hear a baby screaming, you can place plugs in your ears and the sound goes away.  Yet, if the baby is really in trouble, you haven’t solved the real problem.

It seems that this is where we are with our current management of shock.  Because shock is often associated with a low blood pressure, physicians are increasingly giving vasoconstrictors to make the blood pressure go up.  But does this actually fix the problem?  Is fixing the blood pressure really fixing shock?

It’s an actual paradox.  Definitions of shock published in textbooks and the medical literature over the past 50 years attest to the fact that shock is the presence of poor tissue perfusion, or energy supply, and is not the result of low blood pressure (although a low blood pressure is often associated with shock).  Also, low blood pressure is understood not to represent shock (although it can warn of its possible presence).  Yet, clinical practice seems focused on improving the blood pressure, even if the treatment makes no mechanistic sense.  A recent study compared two vasopressors, dopamine and norepinephrine, and determined that there was no significant difference in outcomes between the two.  While the authors recognized that their patients had shock due to various causes, it seems that vasoconstriction was considered a key therapy because the blood pressures was raised.  In fact, about a third of the patients had hypovolemic or cardiogenic shock, in which vasoconstrictors are usually considered potentially harmful.  Of course, what was lacking in the study was a control group of patients who received no vasoconstrictors at all, but actually had an individualized hemodynamic analysis with the application of appropriate remedies to correct the circulatory defects.  The lack of a control group implies that everyone is sold (with no real evidence) on the concept of raising the blood pressure at all costs, even though it may adversely affect tissue perfusion.  It will require a paradigm shift in medicine to redirect the mechanistic approach to physiologic resuscitation.

Judgment and Experience

July 28, 2010

Good judgment, it is said, comes from experience.  And experience, in turn, comes from bad judgment.

In a true sense, this principle is how we trained our surgical residents for nearly a century.  They were primarily responsible for the patient’s care, including their surgical procedures.  Experienced attending surgeons were available to assist or guide the resident when the need arose.  But mostly, senior residents taught and supervised junior residents, and junior residents taught and supervised interns.  Because of their responsibilities, surgical residents had to make decisions they could justify, and they were accountable for any poor decisions they made.

This all changed between the 1980s and the 1990s when Medicare refused to pay attending surgeons who weren’t physically present during operations.  This produced a paradigm shift in medicine, providing significant cost savings for the Medicare program.  Yet, it crippled the educational value of surgical residencies.  By making the attending physician be physically present for every operation, the resident is no longer in a position to make independent decisions.  In most cases, the attending calls all the shots, leaving the resident as an assistant or even an observer.  Surgical decision-making and judgment are no longer being learned to the same degree.

Consequently, many surgical residents finish their training feeling unqualified to practice on their own.  Some opt for fellowships in order to gain more experience in a supervised environment.  Once they have completed their training and enter practice, however, they often state that it took them 5 to 7 years of practice – which is, coincidentally, the range of time spent in most residencies – before they felt entirely comfortable as an independent operator.  It’s really too bad they couldn’t get that training during their residency, where they had actual supervision and support.

It’s true that we’re talking about patient care.  And in previous eras, we let incompletely trained and experienced physicians manage those patients more independently.  But how should these young physicians gain their experience?  How else can they acquire the ability to exercise good judgment under pressure?  Should this happen when they are out on their own, trying to build a practice and recruit patients when they have no structured supervision?  Or should it be during their residency program where those financial pressures do not exist and attending physicians are built-in as available resources to supervise and guide them?

Spray Painting Them Pink

July 20, 2010

The paradigm shift in medicine is to fix the underlying problem, not its manifestation. 

When the heart stops, oxygen supply to the tissues is cut so drastically that cells have to try to survive without oxygen, a process known as anaerobic metabolism.  One of the consequences of this is an accumulation of hydrogen ions from the hydrolysis of ATP exceeding the rate of ATP production, producing a metabolic acidosis.  This can be demonstrated on an arterial blood gas sample as a low pH (i.e., less than 7.40) and a negative base excess, otherwise known as a base deficit.  It has been demonstrated that the lower the pH and the more severe the base deficit, the worse the patient will do, with death being the likely consequence for those with the most abnormal values.

We were first able to detect this clinically after the invention and development of blood gas analyzers in the late 1950s by John Severinghaus.  Physicians began to note this acidosis in the early 1960s when cardiopulmonary resuscitation (CPR) was introduced.  It was also noted that if sodium bicarbonate, a base, was administered, the acidosis could be reversed.  For decades, it became standard practice to administer “bicarb” as a routine during CPR.  The justification was that acidosis could impair cardiac performance and its response to catecholamines such as epinephrine (adrenaline).  Because of this, clinicians were instructed to correct acidosis with sodium bicarbonate before giving epinephrine during CPR.

Of course, this was all theoretical because we weren’t directly measuring cardiac performance or catecholamine responsiveness.  That was all below our radar.  In fact, there are a number of studies and reviews spanning back several years that attest to the fact that such bicarbonate is, at best, of little use or actually harmful because of the severe acidosis it can produce inside the cells (which is definitely beneath our radar).

The simplistic approach was to fix a number that was actually the result of a problem rather than the problem itself.  Correcting acidosis by giving bicarb is as effective as treating a suffocating cyanotic (blue) patient by spray painting them pink.