The most important thing you can do as a psychiatrist...

 

I won’t build the suspense.  The most important thing you can do as a psychiatrist is to be the medical doctor that you were trained to be.  The second-best thing is to be a good if not great psychiatrist.

I had those thoughts today after reading about a case of misdiagnosed panic disorder (1).  The patient was an athletic 30 yr old women who reported episodic panic attacks, palpitations, light headedness, and shortness of breath.  A Cardiology evaluation was negative.  That was not too surprising since she was asymptomatic during the testing, but given the final diagnosis I would have expected a subtle baseline ECG change.  She was treated with a selective serotonin reuptake inhibitor for presumed panic attacks by her primary care physician.  She is seen in the Emergency Department and an ECG shows an irregular, rapid, wide QRS complex, tachycardia and her usual symptoms. A shortened PR interval with a delta (preexcitation) wave is noted. The entire case description with the associated diagnostic reasoning can be located at this link (1). I am not sure that readers can access it without an account.

The case is an excellent example of the real task of being a psychiatrist. The usual dialogue about what psychiatrists do is typically restricted to criteria in the Diagnostic and Statistical Manual (DSM).  There is a lot of confusion about the importance of the DSM and what it means for psychiatric practice. For example, the popular stereotype is that psychiatrists just sit around and estimate whether people “meet criteria” for a DSM diagnosis and then prescribe an indicated medication.  Life as a psychiatrist is not that simple.  The unique problems of the person in front of you cannot be captured by a crude system of classification.

Using this case as a backdrop, I need to know as much medical detail about this young woman as possible.  More details about the onset of symptoms and associated symptoms. More details about her baseline physical health, associated symptoms, and any cardiology consultation and testing that has occurred.  If I am on the same electronic health record system, I am pulling all of that up including her vital signs over time, lab testing, and cardiac testing.  I am looking at each ECG tracing.  I need to know her detailed family history for cardiac disease, arrhythmias, and sudden cardiac death. How much alcohol, tobacco, and caffeine does she typically use?  Is she using any stimulants?  Does she have an intercurrent illness that could affect her heart rate?

In the next few minutes, I need to be checking her vital signs especially her heart rate and rhythm, respiratory rate, and doing a rapid cardiopulmonary exam. My first decision point is whether she is in a medical emergency or not. This is not always as clear cut as this case where the discussant points out that the patient is treated using the American Heart Association (AHA) Advanced Cardiac Life Support ( ACLS) algorithm and needs electrical or medical cardioversion.

That is where things get tough for a psychiatrist.  Setting is a significant issue.  If I am working in an acute care setting in a hospital – I typically have plenty of back up.  Hospitalists services generally run codes or even have a team for acute care that does not involve codes and I could get them there in a few minutes.  At the other end of the spectrum -  I have worked in a community mental health center with absolutely no access to ECGs and no equipment for cardioversion.  In that case – 911 needs to be called and all medical staff in the facility should be able to perform basic cardiopulmonary resuscitation. 

The main work in this situation is recognizing the medical emergency and getting the patient to the correct setting where she can be stabilized. It is not always black and white.  This patient was eventually diagnosed with atrial fibrillation and Wolf-Parkinson White (WPW) syndrome.  Atrial fibrillation was probably the most frequent cardiac diagnosis that I made as a psychiatrist.  Most people who had it were not aware of it. I happened to pick it up because I noticed an irregularly irregular pulse when checking their vital signs and a pulse deficit on physical exam. It was almost always in a range where the heart rate was not a big problem.  In some cases, it was partially treated by a rate controlling medication like a beta-blocker or calcium channel blocker.  I could typically call the patient’s primary care physician and get them in for a comprehensive evaluation of the problem.  I would have to send some patients to the emergency department or urgent care.     

The issue of cardiac related anxiety is a very interesting issue. Cardiac symptoms can be an associated symptom of anxiety, panic, and other affects like anger.  The symptoms can arise as a sensory phenomenon due to an awareness that the heart is “pounding” or “beating out of my chest”.  Both of those descriptions are very common in people with panic attacks. The sequence of events and what is causing the cardiac phenomenon are wide ranging from an intrinsic cardiac problem to an imbalance in the sympathetic and parasympathetic innervation of the heart. Some electrophysiological experts think that at least some atrial fibrillation is due to overactivity of both autonomic systems.  Even in the absence of a sustained arrhythmias – the autonomic effects can result in premature atrial contractions, premature ventricular contractions, and sustained sinus tachycardia.

There are many other cardiac emergencies that occur in psychiatric settings. I was asked to see an acutely manic woman who was 85 years old.  She was extremely agitated and shouting that her chest hurt.  I was able to get a stat ECG that showed she was having a myocardial infarction and got her transferred to the coronary care unit.  In another case – I was told that a 70-year-old woman was “delusional” about her abdomen.  She clearly had a belief that there were supernatural forces causing her abdominal discomfort.  At the same time, she had a pulsatile mass in her lower abdomen and an abdominal aortic aneurysm on ultrasound.  Both patients survived with timely intervention.

I was a quality reviewer for many years and that job involved reviewing potential quality problems associated with inpatient hospitalizations.  One of those reviews was a patient who was hyperventilating.  He was diagnosed with panic attacks and treated with behavior therapy that did not seem to be effective.  As his condition worsened, he was eventually diagnosed with an acute pulmonary embolism. Since that review, I have seen many ambulatory patients who were short of breath for days due to pulmonary emboli and are not seen in a setting where they can be diagnosed and treated.

There are many more medical problems that crop up in psychiatric outpatients and inpatients that cannot be missed.  They can present as a possible psychiatric disorder and the potentially fatal nature of many mean they cannot be missed.  Many settings are set up to give the appearance that an emergency room physician, hospitalist, primary care physician, or physician extender is medically clearing these patients and that is not the case.  Most frequently that is because the time course of the condition is erratic or communication with a psychiatrist at a more detailed level is necessary.  The only assurance that these patients have no acute problems is if they are acutely symptomatic when they are screened or seeing a psychiatrist who can communicate with them, has no biases against them, and who knows the difference between a medical and a psychiatric problem.

To be very clear, I am not suggesting that psychiatrists initiate care for life threatening medical problems.  They do need to know if there are undiagnosed acute or chronic medical problems and how fast they need to be addressed. They need to be able to recognize the medical causes of signs and symptoms that can be misinterpreted as psychiatric.  They also need to recognize and manage the associated systems problems that in my experience are primarily countertransference driven.  Let me provide a clear example of what I mean.  I was working in an acute care setting and came across a patient leaning against a pool table. He was acutely short of breath, somnolent, and had a history of valvular heart disease. On exam, he was in congestive heart failure. I placed a call to the medical consultant and was told that I should start an IV line and manage the patient myself on an acute care psychiatric unit.  When I suggested that he needed transfer to medicine – I was met with the comment: “Well you know how to start an IV don’t you?”

I most certainly have started hundreds of IVs, but that is not the issue.  My patient had an acute medical problem that needed both medical and nursing expertise to manage in a more medical acute care setting than a psychiatric unit.  I eventually contacted the Chief of Medicine and got the patient transferred where he was subsequently in an ICU setting.  Ideally acute care psychiatrists today can develop good relationships with hospitalists for these kinds of transitions.  The best way to do that is by letting them know you have made a medical assessment and have a good indication for transfer.

Being a good if not great psychiatrist is hard work. My most significant worry was missing a major medical problem and not getting adequate intervention.  That is just the first step. The next steps are a psychiatric formulation, diagnosis, and treatment plan that incorporates state of the art communication and relationship building with the patient.  Hopefully that is followed by a long period of seeing the patient, helping them meet their goals, and providing medical diagnosis and follow up as needed.  In today’s world that is often occurring in a rationed suboptimal environment, overburdened by businesses rationing of both care and medication for profit.

My hat is off to the psychiatrists who are doing this work and probably working way too hard in 2025.

Happy New Year!

 

George Dawson, MD, DFAPA

Supplementary 1:  There is no doubt that I have practiced in settings where there was a high level of concurrent medical and in some cases surgical illness.  There is also no doubt that it was a conscious decision on my part to practice in those settings.  That undoubtedly sharpened my focus on making sure that I had the skills necessary to provide adequate care to those populations. It may be possible to cleanly partition psychiatric work from the rest of medicine but I have not seen that happen for some of the reasons cited in the above essay.  The training of psychiatrists in the past has had a variable relationship with medicine - at one point going to the extreme that much of the medical internship was eliminated.  The best advice I got in medical school was not to use elective time for additional psychiatry rotations because I would be doing psychiatry the rest of my life. I took neurology, neurosurgery, nephrology, endocrinology, cardiology, allergy and immunology, and infectious disease rotations instead. That initial training worked well over my years of practice and I don't regret it.

Some may question the emphasis in this post on the importance of not missing concurrent medical diagnoses and I would offer these additional observations.  Many patients seeing psychiatrists consider them to be their primary care physicians. That should not deter a psychiatrist from clarifying their role, but the fact that psychiatrist is probably seeing the patient much more often than the primary care physician is often a useful reality.  I have called primary care physicians to report what I consider to be an exacerbation of the patient's underlying medical problem.  That collaboration can get more timely care for acute or chronic medical problems.  I have also had the experience hearing from a person that a psychiatrist diagnosed their medical problem when nobody else did.  Many of these scenarios degenerate into who is the better physician.  The focus needs to be on what the patient needs rather than what the physician needs.  Not ignoring or missing a patient's underlying medical problems is a large part of that personalized care. 

 

Reference:

 1:  Hemingway TJ.  An athletic patient who thinks she has panic attacks.  Medscape December 17, 2024 (accessed on December 31, 2024):  https://reference.medscape.com/viewarticle/858516_6

Waiting List Mortality? - An Example of Nocturnal Panic Attacks

Any PubMed search on waiting list mortality will produce a long list of articles on mortality that occurs on transplantation waiting lists and all of the associated ethical and logistic problems.  I could not locate any work done on waiting lists to get in to see psychiatrists, primary care physicians, or specialists.  To an extent, waiting list mortality is expected and some of the risk factors (increasing age, significant chronic medical illnesses, high  risk medications) is undoubtedly predictive.  But what about the person who calls in and describes a clear cut problem that is misclassified and the error is potentially life threatening.  I have picked up a few of these problems in psychiatric clinics where the ultimate emergency diagnosis was unrelated to the reason for the appointment.  Seeing a patient who is white as a ghost, complaining of coffee ground emesis, and determining the hemoglobin to be 7 is one of many examples.  Anxiety was the reason for that appointment.  I have found a number of acutely anemic patients due to blood loss with complaints about anxiety, shortness of breath, and panic attacks  who really needed blood transfusions.

I have a more concrete recent example that is much more common in psychiatric practice, especially if a significant number of patients with alcohol and benzodiazepine problems are being seen.  That is the problem of nocturnal anxiety with panic attacks.  To make it a little more interesting, let's say our hypothetical patient is 55 years old, has a history of paroxysmal atrial fibrillation (2 brief episodes), does not take an anticoagulant, and takes flecainide to prevent atrial fibrillation and metroprolol to prevent palpitations.  He has obstructive sleep apnea and is on APAP.  His AHI is 3.5 or less on any given night.  He has never has an electrophysiological study. Stress test and echocardiogram were both negative.  He has had to taper the metoprolol over a period of about 5 years from 25 mg BID to 3.125 mg daily due to lower and lower BP.  Suddenly the patient is noticing palpitations at night.  They seem to occur at the end of REM type dreams and do not seem to correlate with the emotional content of the dream.  The awakenings with palpitations typically occur at 4AM.  He has not had a panic attacks in 35 years.  He does not drink or use benzodiazepines.  To terminate the palpitations he gets up and drinks a large glass of water or walks around and they resolve in about 10 - 20 seconds.  He gets a small single lead ECG device that reads one aberrant beat as "occasional PVC".

He gets in to see his primary care MD in a couple of days.  His exam is normal and he is in sinus rhythm.  No aberrant heart beats are noted by his internist.  Electrolytes and magnesium level are normal.  He goes home that night and the palpitations continue.  He calls his sleep medicine physicians who tells him to start with his Cardiologist.  He calls the Cardiologist who concludes that "these are beats that originate in the lower chambers of the heart and there is nothing to be concerned about." Despite the acute change in symptoms that is waking him up on a nightly basis - no further testing, examination. or diagnosis is offered.  Frustrated - he calls a referral center and schedules a sleep study in about 2 1/2 months.

Is there something else that could have happened in this case? If you happen to be this man's psychiatrist - like I am in many of these cases of sleep related symptoms what is the differential diagnosis and what else can be done.  A reasonable differential diagnosis of these palpitations might consider the following list of conditions.

 

Night time palpitations – differential diagnosis:
1.	Sleep  terrors
2.	Nightmares
3.	Nocturnal panic attacks
4.	Alcohol or sedative hypnotic withdrawal
5.	Stimulant or hallucinogen intoxication
6.	Cardiac arrhythmia Tachyarrhythmias Ventricular arrhythmias Supraventricular arrhythmias Conduction delay arrhythmias

In considering the list. there are some useful clinical features.  Sleep terrors are rare in adults.  They typically occur in the first half of the night.  The patient suddenly arouses from sleep.  They may  scream.  They have intense sympathetic output including diaphoresis, flushing, tachycardia, tachypnea,and mydriasis.  They may appear to be disoriented.  Nightmares are typically dreams with negative emotional content that occur with awakening from non-REM sleep.  Sympathetic arousal is not as prominent.  Nocturnal panic attacks (NP) can occur in people with daytime panic attacks (DP) or as a separate entity.  People with combined DP/NP had more symptom severity. Palpitations are a feature of panic attacks.  In a recent study (2) the authors also rule out associated disorders  like substance use problems and obstructive sleep apnea by exclusion and testing.  The pure NP group were predominately male, had a childhood history of sleep terrors, and were more likely to have respiratory symptoms (choking sensations) despite a lower overall symptom severity score than the DP/NP group suggesting a more common mechanism with night terrors.  Because of the similarity between nocturnal panic attacks, sleep terrors, dream anxiety attacks and nocturnal seizures some authors encourage "extreme caution" in making the diagnosis (3).

The alcohol and substance intoxication and withdrawal states may be less obvious outside of treatment setting specializing in these disorders.  Patients with these problems may not disclose the full extent of use.  In the proper context, discontinuation of both cannabis and hallucinogens like LSD occurs due to increasing anxiety and panic attacks.  Alcohol and sedative hypnotic withdrawal can cause prominent sympathetic symptoms including night sweats, tachycardia, and panic attacks.  Those symptoms typically resolve with treatment of the underlying withdrawal syndrome.  In some cases the anxiety and panic attacks persist and require additional treatment.

Pure cardiac symptoms associated with sleep can be confusing.  The patient is aroused and may notice the arrhythmia.  The question becomes is the arrhythmia secondary to anxiety and sympathetic arousal or is the anxiety secondary to the chest sensation?  In the case of a patient with known sleep apnea and atrial fibrillation here are several possible causes including breakthrough atrial fibrillation at the time of the awakening.  In this case there was a crude monopolar tracing that showed a ventricular premature beat (VPB).  VPBs are commonly associated with anxiety, but is it enough to count on the patient capturing the event by getting up out of bed and holding an inexpensive device to his chest?  Patient with sleep disordered breathing are at increased risk from nocturnal death (midnight to 6AM) and one of the mechanisms may be arrhythmia from the cardiac effects of sleep disordered breathing. 

The patient in this case is very much alive and functioning at a high level.  He still has the nocturnal palpitations but is less anxious about them because they are now intermittent and always seem to resolve in a short period of time. If he thinks about it for nay length of time the question that comes up is: "Why now?"  He hopes that the referral center will have the answer to that question.

  I don't have any outcome or ready solutions to this problem. In many ways it highlights a potential quality problem in the high tech American healthcare system.  Here we have a patient who is fairly compulsive about his own health care.  He has a primary care physician who he saw and contacted both his specialists.  When there was no answer, he contacted a referral center and set up an appointment 2 1/2 months out into the future.  What will happen while he is on that waiting list is a probability statement with a series of unknown probabilities.  What is disappointing from my perspective as a physician trained in the 1980s is that at some point - the rigorous intellectual approach to patients problems has fallen by the wayside in favor of rationing.  We are no longer in pursuit of a diagnosis that might make a difference.  We are satisfied with saying "I checked off all of the boxes and I didn't see anything".  It's a 21st century variation of the old joke: "The operation was a success but unfortunately the patient died."

I am very interested in what the cost of this approach is in terms of human life and additional comorbidity.  I think that what happens to people on these waiting lists (compared to controls) is where the emphasis should be place and not on how fast patients should be discharged and not readmitted to hospitals.  Despite all of the press about unnecessary tests and the risks associated with those tests, the commonest errors I see that result in patient injury is missing the obvious diagnosis and not doing the appropriate tests.     

There is something wrong with a health care system when a psychiatrist cares more about these problems than the physicians running the system.               


George Dawson, MD, DFAPA


References:

1:  Selim BJ, Koo BB, Qin L, et al. The Association between Nocturnal Cardiac Arrhythmias and Sleep-Disordered Breathing: The DREAM Study. Journal of Clinical Sleep Medicine : JCSM : Official Publication of the American Academy of Sleep Medicine. 2016;12(6):829-837. doi:10.5664/jcsm.5880.

2: Nakamura M, Sugiura T, Nishida S, Komada Y, Inoue Y. Is nocturnal panic adistinct disease category? Comparison of clinical characteristics among patients with primary nocturnal panic, daytime panic, and coexistence of nocturnal and daytime panic. J Clin Sleep Med. 2013 May 15;9(5):461-7. doi: 10.5664/jcsm.2666. PubMed PMID: 23674937.

3: Shouse M, Mahowald MW.  Epilepsy, sleep, and sleep disorders.  in: Kryger MH, Roth T, Dement WC. Principles and Practice of Sleep Medicine. 5th ed. St. Louis, Missouri. Elsevier Sanders, 2011: 1048-1063.