Monday, September 29, 2014

Letting Go

Sometimes, death is easier for the patient than for the clinician. I met a wonderful 90 year old vet with a turn-of-the-century name (omitted for privacy reasons) who had metastatic end-stage cancer. He was receiving palliative care, but because of worsening pain, he decided to undergo a palliative surgical procedure. Although we sometimes think surgery is at odds with palliative care, there are times when the intent of surgery is not curative, but simply to make a patient more comfortable at the end of his life. This of course must be weighed against the risks and costs of a surgery so it's a rare event but certainly an appropriate one in some circumstances.

I met the gentleman in the ICU after the case. He was fairly hypotensive and so we resuscitated him, supported his blood pressure, and extubated him from mechanical ventilation. Boy, did he have a personality. He learned all our names (insisting on examining our badges), gave us sage advice ("To get to 90, have a glass of wine every day), and told us delightful stories from his nine decades of life. We grew pretty attached to him. When we talked to him about his "code status," he was quite clear - no heroic measures to resuscitate his heart and he didn't want us to "put that stick down his throat again." His daughter agreed wholly with his wishes. He told us he wished to have his body donated to a medical school. He had lived a wonderful life and made plans for his death. But day after day, he seemed to turn around, requiring fewer blood pressure medications and recovering from surgery slowly.

Unfortunately, at 90 with metastatic cancer, recent surgery, and a host of other factors, he had many reasons to have sudden decompensation. Twice, he had an arrhythmia that dropped his blood pressures and rendered him unconscious. The first time, medications brought him back. He returned to his usual personality, playfully bossing the nurses around. The second time, a completely unexpected cardiac arrest, was not reversible.

Surprisingly, this death affected he medical team more than anyone else. The daughter understood completely and accepted his passing. We, on the other hand, wondered what we could have done differently. Was there a medication we could have started earlier? A test we could have done? A quicker response to his arrhythmia? We wanted so much to have saved him because his personality was so endearing. He was so human to us, the epitome of happiness at age 90, and his death was a blow to ideals.

We had to remind ourselves that this was what he wanted, what was best for him. Instead of a prolonged death of suffering, he simply lost consciousness and passed. His wishes were upheld; he didn't have painful chest compressions or an undignified breath tube shoved in. He spent the last of his days joking with his daughter, telling me about his hobbies (he spent several hours a day on the computer corresponding), and assuring us he was happy and comfortable. Then, within minutes, he simply crossed the bar. In modern medicine, this was probably the most dignified, wholly appropriate, and somewhat rare kind of death.

Saturday, September 27, 2014

Knowing Your Environment

Every hospital is different, and getting to know your hospital is a big part of being an ICU fellow. When we make decisions about a patient's "disposition" - that is, what level of care a patient needs - we have to be aware of our hospital's capability. I mention this because the VA is quite different than Stanford. The ICU at the VA often has patients whose acuity isn't all that high simply because they have specific nursing needs that only the ICU delivers such as hourly vascular checks. Patients that may go to the floor at other hospitals come to the unit at the VA. This also occasionally happens because the floor's ability to take care of sick patients is quite limited at the VA. I've often taken patient's to the ICU who might have stayed on the floor at a different hospital. This has a lot of implications, including cost considerations; an ICU bed uses a lot of resources and costs quite a bit, so having a system where a nonacute patient may take up slots might not be the most efficient. But it can also create an odd mix of patients in the unit. A patient may be moribund, on multiple intravenous pressors, mechanical ventilation, and dialysis, while his neighbor might be chatting, reading the morning newspaper, having breakfast, and simply having his pulses checked hourly. It's an odd unit in that respect. At times, I feel like I'm running a medicine ward service.

The other aspect of the VA is that it runs pretty close to capacity all the time. When patients come into the emergency department, I occasionally struggle to find them a bed. We wait for days to transfer patients out of the ICU. This, too, increases the cost of health care; we have patients who are appropriate for the floor taking up ICU resources simply because no other bed is available. Perhaps because the VA is not a profit-focused hospital, this is less important. However, considering our overall national expenditure on health care, we have to keep such things in mind.

Wednesday, September 24, 2014


The VA ICU is a special place for me for lots of reasons. It is where I started the very beginning of internship year, when my white coat was clean and pressed, when my transfer summaries were gregarious novels, when I was bright-eyed and bushy-tailed. Surprisingly (or perhaps not), it hasn't changed too much. This time returning, I recognized many of the same nurses, argued with the same surgeons, dealt with the same frustrations. But there is a small feeling of recognition, as if returning home.

When I started internship year, I had no sense of perspective, and the VA in Palo Alto felt like a large and intimidating place. As I was sent out to find the spinal cord unit, the preoperative clinic, the emergency department, I marveled at how large the campus was. Now it seems like a tiny microcosm. I used to find the 15 bed ICU intimidating, but now, it's a breeze. I've figured out all the processes, realized how I fit in. And I've seen my peers go through this process too. One of my cointerns is now a medicine attending at the VA and we shared a couple patients together.

With only 15 beds, it is good training grounds for interns. Some days, I am frustrated by how long rounds take, but I remember that I was there once, that this is hard, and that interns have a steep learning curve to surmount. But it also gives me great opportunities for teaching, and it's so refreshing to have time to explain my thought process, discuss our patients, challenge misconceptions. We also have residents and nurse practitioners who help supervise the interns, and they are ever a source of learning, amusement, and assistance. My job was made so much easier by my brilliant residents who kept the interns in line, taught them so much, made sure all the checkboxes in patient care were completed. This structure with interns and mid-levels gave me the opportunity to play junior attending, running rounds, making decisions, and overseeing big picture plans. Having this leadership role was really important for me to gain confidence and nudge me out of my normally introverted personality. After CVICU where I played a resident role, this was refreshing change.

Monday, September 22, 2014

Finishing Up Round Two in the CVICU

As I reflect on my last few days on the cardiovascular ICU, I remember great relief and also strangely, a little longing. It is the feeling of being forced to do something I wasn't sure I could handle, and then coming out intact at the end. There was a period of time in there where I worked 17 days in a row that I had this inexplicable mix of exhaustion and accomplishment. Each day we spent 12-16 hours in the hospital, and after two and a half weeks of that, I knew every single detail about every single patient. I understood their hearts so intimately, could tell you how the patients responded to each hemodynamic change, could recount the story of recovery after surgery. This is medicine as was practiced generations ago, when the word "housestaff" and "resident" were literally true. The hospital was my home, the patients, nurses, surgeons my family. I learned so much and felt invincible. I took the best parking spot when I arrived at 4AM, learned the quickest routes to the cafeteria, found stashes of snacks for ailing residents and fellows. I got to know which family members slept in their loved ones' rooms, which nurses took extra night shifts, which anesthesia techs staffed the graveyard shifts. I loved that feeling.

But it was also a month where I seldom saw my wife, abandoned my blogs, forgot about cooking, became derelict in emails, became lost to follow-up to my friends and family. One of my friends' parents refers to her OBGYN residency as the "dark ages," a gap of four years where she knows nothing of current events, movies, books, pop stars, friends' life events, elections, politics. It is a sacrifice physicians generations ago made, and something we seldom do now. After going through this month, I understand why such a period of clinical immersion is so transformative but also how it can wreak havoc on any identity outside being a physician. I am glad I went through my CVICU month. I am glad it is done.

Sunday, September 21, 2014

Root Cause Analysis

When a really bad outcome happens, the hospital undergoes a process called root cause analysis to figure out what went wrong. It's easy to point fingers at superficial or proximal elements that led to the incident, but this formal method of evaluation assures us that all contributions to a clinical event are identified.

When a root cause analysis happens, it's a big deal, committees upon committees. In this instance, a patient passed away within days of receiving a transplant. Why did this happen? Could we have avoided it? Were there misses with the pre-operative medical management, the surgical technique, or the anesthetic? What could we have done post-operatively to prevent this devastating outcome? With the scarcity of organs and donors, this was the kind of incident that really forced us to review our processes. With all the stakeholders involved - the medical transplant team, the surgeons, the medical intensive care team, the cardiothoracic critical care team, the nurses, the operating room staff, the transplant coordinators - we really probed into every possible contributor to the patient's death. I hope that this ultimately makes future transplants safer for patients.

We do a lot of transplant surgeries, and I've been in the operating room on the anesthesia side, in the pre-operative evaluation on the medical ICU side, and in the post-operative management on the CVICU side. We have so many successes; patients who have been living on home oxygen or home infusion pumps for years who walk out of the hospital. But it is the sadder, less successful outcomes that help us improve our clinical care. I wish that every patient had a perfect transplant course, but when it doesn't happen, it is our responsibility to scrutinize and fix things. I write this blog not to point out our deficiencies, but to emphasize our constant, ongoing commitment to improving patient care. Every institution has incidents like these, but I think it is important for me to be open and talk about all our stakeholders are committed to processes like root cause analysis.

Wednesday, September 17, 2014

No Benign Procedure

A 20 year old considering a career in professional athletics has a benign congenital heart defect. It causes him no symptoms on a day-to-day basis but it's felt by his coaches and trainers that he should have it repaired. The procedure ought to be a simple one; in fact, the bypass time is only half an hour. He is young, and no one expects any complications. After all, we do open heart surgery on octogenarians. His surgery appears to be uneventful and he comes to my care in the CVICU post-operatively.

We extubate him and wean off his cardiovascular drips. He's meeting all our goals and parameters, but upon working with physical therapy, he notices something odd. He is slightly weak on one side. We rush to scan his head, and there's evidence of a completed stroke, likely from the OR. We are all devastated. In getting an elective heart surgery for a career, a complication makes it impossible for him to pursue that career. He will be fine for everyday activities; he's improved with physical therapy and he was quite strong to begin with. But he will never play professional sports again.

This was one of the saddest cases I had in the CVICU. We all know that complications are a risk of any surgery, but it's easy to assume they would never happen to a healthy young patient. To see someone with such potential lose so much hope - my eyes filled with tears in talking to him. It is a grave reminder that everything we do in medicine has risk, and we should never undertake any procedure lightly. I am glad I am not a surgeon, but as an anesthesiologist and intensivist, I will have similar circumstances and situations. Reminders like this make our responsibilities as a physician feel so much more raw and poignant.

Sunday, September 14, 2014

If I Could Make a Living by Writing

I could never do spine surgery, but a part of me wishes I could be a writer. Image shown under Fair Use, from by Jessica Hagy.

Saturday, September 13, 2014

Observations in the CVICU

No one has actually taught me this, and I'm not even sure it's true, but anecdote sometimes drives medicine. To me, it seems that after cardiac surgery, patients either fly through recovery or they linger with complication after complication; there isn't a middle ground. I almost want to do a study on the length of stay in the CVICU; I'd guess there'd be a bimodal distribution with a bunch of patients leaving 2 to 3 days after surgery and a bunch leaving around 2 weeks. I'm not sure why this, but I have a few guesses. Perhaps run-of-the-mill surgeries - a routine CABG or valve replacement - represent the quick, reliable in-and-out ICU stays and the emergent cases, complex referrals, and high-risk surgeries buy a longer stay. Or perhaps patients receiving cardiac surgery have such little reserve that any complication will set their course back by weeks. In any case, this seems to be the type of observation that an administrator might want to know because the long-term players tie up critical care beds and cost the hospital and the health care system a lot of money.

Tuesday, September 09, 2014


There are some events that are simply catastrophic, events I consider non-survivable. Even though sometimes we can tide patients through the acute phase, I worry that these disasters set the patient back so much that recovery is impossible. It's a terrifying realization, really, because sometimes you feel that you have a patient who has no hope for survival. Sometimes you just hope that your intuition is wrong.

A patient with alcohol related heart failure receives a left ventricular assist device. The LVAD is a continuous pump that assists the heart by sitting in the left ventricle and pushing blood out to the aorta. Post-operatively, these patients are incredibly tenuous because only the left heart has assistance; the right heart remains in failure. They often stay in the CVICU for weeks while we tend to the right heart and address all the usual ICU complications. These patients seem to develop more ICU-related morbidity than standard patients, probably due to their tenuous hemodynamics and implanted hardware. It often takes us weeks to wean epinephrine, dopamine, milrinone, vasopressin, inhaled nitric oxide. And during this time, we struggle with ICU delirium, pain management, pulmonary hypertension, infections, ileus, renal failure, and glucose management.

This patient with an LVAD has had steady progress for over a month. We've been actively diuresing him to offload his right heart, but an unfortunate side effect of drying him out is severe constipation. His stool has become a cement sludge, and though you wouldn't think bowel movements are all that important in the CVICU, this story taught me differently. He was being evaluated daily by general surgery and medical GI consultants and had received several colonoscopic decompressions. I'd never seen anyone that happy to get a colonoscopy; he didn't want any sedation and enjoyed watching his own stool be disimpacted. Every day, we made a little progress with getting his bowels moving, weaning his oxygen, titrating down his inotropic drips.

Then one day, I got called because of severe abdominal pain. When I saw him, I knew something bad was brewing, and in fact, I can remember the very thought: this is what's going to kill him. His abdomen had peritoneal signs, indicating that he had perforated his bowel. I got an abdominal X-ray and as I watched them position him for it, I knew what it would show. A rim of gas was seen in his belly. We called general surgery, and even though he was at incredibly high risk - 3 vasopressors and a fresh LVAD - they took him to the OR.

That first night after he got back tested all my anesthetic and critical care abilities. He had high abdominal pressures concerning for abdominal compartment syndrome. He was maximized on four vasopressors, two of them at continuous code doses. His ventilation was poor because all the irritaiton in the belly had caused swelling, making it hard for us to deliver deep breaths. And the surgeons weren't confident about their bowel resection; they said that his intestines fell apart in their hands as they operated (a result of his vasopressors). The biggest challenge was fluid management. After large bowel surgery, patients become very dehydrated, and so he got a substantial amount of blood product and fluids. However, his right heart could not tolerate that sort of load. If his right heart failed, he would die.

We sat at his bedside all night, pushing medications, starting paralysis, doing serial echocardiograms, managing continuous renal replacement therapy (he had gone into kidney failure during this). And we managed to tide him through the acute catastrophe. But with ischemic bowel, raging peritoneal infection, right heart failure, kidney failure, prolonged paralysis and steroids, minimal nutrition, and increased hemodynamic support, I worry that this catastrophic event set him back a whole lot, and likely too much. Now I am in a hard place because we are doing everything we can to save this guy, but deep in my heart, I feel that he will never realistically recover enough to make it out of the ICU. I hope fervently I am wrong, but after a few years in medicine, I realize there are some catastrophic events where modern medicine can weather someone through a little bit, but not enough.

Saturday, September 06, 2014


Physicians often use the word "protoplasm" to mean a patient's overall health and fragility. Protoplasm makes a big difference. A 70 year old with hypertension, hyperlipidemia, coronary artery disease, aortic stenosis, COPD, end stage renal disease, peptic ulcer disease, diabetes, and Parkinson's fares a lot worse from a flu than a 70 year old who walks his dog five miles a day and has no other significant medical problems. Someone who is bedbound, wheelchair-bound, or limited in mobility has a higher risk for complications than someone who ambulates on his own. A patient who is cachectic, who has little reserve, has a higher likelihood of dying from any physiologic perturbation.

All this is obvious, but some experiences in the CVICU really place a spotlight on "protoplasm." I was shocked to hear that one of our surgeons did a "triple valve" on an eighty-year-old. She had an aortic valve replacement, a mitral valve replacement, and a tricuspid ring. The surgeon was extremely skilled and the bypass time was fairly reasonable. But the patient's age, pre-existing pulmonary hypertension, baseline kidney disease, and frailty made the post-operative course a nightmare. We spent days weaning our inhaled epoprostenol and pressors. We had to be ginger not to provoke the right heart as she teetered on right heart failure. We managed to extubate her, but her poor pulmonary reserve required reintubation the following day. We struggled to get adequate nutrition for her. We could not salvage the kidneys and had to start continuous renal replacement therapy. In retrospect, she was not an appropriate surgical candidate. Although you may be able to operate on anyone, you shouldn't.

The problem in our current system is that pre-operative decision making doesn't include all the players. It is up to the surgeon to realistically set expectations on the recovery course and an anesthesiologist often ensures that things are as optimized as possible. They ensure that the surgery is feasible and that the patient will make it through, but don't always assess how rocky the post-operative course might be. As a critical care physician, I only meet these patients after the fact. And the truth is, for some patients and some surgeries, the stay in the ICU is harder than the surgery or anesthetic, and a patient ought to know what that might look like before committing to it.

Thursday, September 04, 2014

Back in the CVICU

Although one of my last rotations as a resident was in the cardiovascular intensive care unit, I was not gone for long as it was one of my earliest rotations as a fellow. The role of the resident and fellow is pretty much the same, and although the second time on the rotation was a little bit easier and made more sense, it was just as exhausting.

The thing about the CVICU is that the patient acuity is higher than anywhere else in the hospital. Our surgeons offer surgery to patients who have been deemed too high risk elsewhere. We take transfers of patients from all over Northern California and have expanded the procedures we perform. On "TAVR Tuesdays," the day where our surgeons and cardiologists perform minimally invasive aortic valve replacements (TAVR stands for transcatheter aortic valve replacement), we routinely get patients in their 80s and 90s.

All of these factors make post-op management incredibly challenging. I've had patients on six or seven inotrope, vasopressor, or pulmonary hypertension agents: epinephrine, norepinephrine, vasopressin, phenylephrine, methylene blue, inhaled nitric oxide, and epoprostenol. We struggle not only with complex cardiopulmonary physiology - I've seen pulmonary artery pressures 110/50 - but also with sepsis, gastrointestinal bleed, and anuric renal failure. Balloon pumps, ventricular assist devices, and transplants make up half our service. Compared to my previous month, we only had a few patients on ECMO, but the overall service was sick. I had to perform emergency trans-esophageal echocardiograms several times, rush patients back to the OR, and manage rocky intubations and extubations.

On some level, I loved it. I picked ICU medicine and anesthesia for the moments where someone's life depended on a thoughtful but immediate synthesis of an array of complex data. I like the adrenaline, the fear, and the thrill of seeing patients on death's door get better. But overall, the month sapped me. We had some staffing problems which put a lot of strain on the residents and fellows that month, and I really appreciated my colleagues who all put in more time than expected to care for our patients. There was a time when I could fill my life with work, food, sleep, and exhaustion, but when I think of a job in the future, I think I'll want something with less craziness.

Monday, September 01, 2014

Behavior Change

We talk about behavior change as if its impossible. We get teaching about the transtheoretical model where a person must go through various stages before behavior change happens: precontemplation, contemplation, preparation, action, and maintenance. When we meet a smoker of 50 years, we throw up our hands and give up. But should we? Is change really that tough?

Over a year ago, most of the counties around here started enacting laws that required grocery stores to charge for plastic or paper bags. It was a small thing to encourage behavior change. Consumers grumbled. We'd forget to bring our reusable canvas bag and we'd mumble about paying that extra dime or quarter. We wondered what we'd line our trash cans with. But deep down, we understood the environmental principles beneath this change. Now, whenever I go to the grocery store, I and most other customers I see lug along our reusable bags. It's ingrained in us now, and perhaps this small but universal change will help out our landfills some, reduce the plastic we produce and waste.

Similarly, a year ago, our ICU progress notes changed. The decision from the "higher-ups" was to change our notes from "organ-systems based" to "problems based." The details are unimportant, but this translated into a lot more work for the residents. I used to be able to jot out a note in a couple minutes; now, as I click through our electronic medical record to add problem after problem, the notes become much more time-consuming. There was a lot of resistance at first; we didn't understand why this needed to be. But now, a year later, it's how things are done.

To enact change, we'd like the person to be ready for it. We'd like all our smokers to be reading about alternatives, going to counseling groups, and talking to their families and friends about quitting. But in absence of such an idealistic world, we should not give up. We should continue to push for rules, regulations, incentives, and disincentives to nudge those along who need behavior change. I used to think that increasing the tax on cigarettes would not dissuade smokers, but I've come to question that notion. It's true that everything has pros and cons - and a tax like that might have a biased effect on a certain socioeconomic class - but behavior change is paramount for public health, and we should do everything we can to make it happen.