Images

Three weeks ago, I went to see my surgeon, the ob/gyn who performed my hysterectomy last fall. The last time I had seen her was in January, the day my complication was diagnosed. In my mind, though, she had been present throughout my various procedures and recovery, even after the urologists took over my care and our only continued connection was through what I sent to her. I had talked with her once after the urologists placed my drains, and after that, emailed (my preference, a consequence of my lifelong terror of the phone). I had sent her my article during those first weeks, and she emailed back that it described her own experience. Time had separated us more by the time of my visit, but I kept her up to date with plans and surgery reports.

We had become entangled by concern as I worried as much over her experience as my own. So to the extent that my meeting with her last month was a conclusion, I was sad to see her go, and I felt unaccountably lonely that day. This long, shared experience, that wasn’t really shared but perhaps co-imagined, had made me want to cling to that connection. But yet, what to do with it? And the awkward: she is not my colleague; we are not friends.

And yet, as I spoke with her, I found myself wanting to ask questions like a colleague/ mentor/ teacher/ friend more than as a patient. What did you see, what did you do, what (if anything) have you changed? And to explain what I had surmised, technically: that my injury was subtle and would have been hard to detect at the time of surgery, my ureter at first patent, then after 20 days blocked, then, soon after, ruptured. That my best guess is a devascularization injury to the ureter that caused it to deteriorate slowly. That, at the end of surgery before closing, everything might have looked normal.

Long before I had my surgery, I knew that she was a low volume surgery provider. I know the data: that, statistically, lower volume surgeons have higher complication rates. I knew it before my own surgery and thought but didn’t seek to make a change. I probably could have asked to go to the large referral hospital in the same town– I certainly could have withstood the awkwardness of the request. Perhaps I partly thought that the odds were still (and ever) in my favor, as they were. Statistical risks don’t get you far as an individual. I liked her hands.

Why would high-volume surgeons have lower complication rates than low-volume surgeons? Partly it may be the development of the manual, physical skill of surgery. Practicing a fine motor task leads to greater skill, and perhaps in this case the more you practice the greater the skill. But I think also that there is an accumulation of knowledge about anomalies and an abundance and diversity of feedback as one does surgery more often.

In some surgeries and for some complications, the feedback is immediate. This fast feedback enhances learning because the error and the consequence are memorable and easy to connect. These complications may be painful or scary or intense in the moment, but the connection between cause and effect is never in question.

In my case, the feedback was long delayed. By the time I had my ureter reimplantation surgery, suture from the original surgery had dissolved, scar tissue had formed, and the original injury was completely obscured. Even worse, the only way that my surgeon knew about my progress and the only way she saw my surgery report was that I sent these things to her. Otherwise she would have had no follow-up or feedback at all after the complication had been recognized. Without my efforts, her knowledge of my complication would have been limited to the findings on the day of its discovery.

This lack of feedback feels perverse. It is the consequence of privacy laws and complicated, diverse, and poorly interconnected systems in medical care. But it is not the way to support doctors or to improve the quality of care they provide.

Atul Gawande describes the need for a more robust feedback loop in medical care in this Freakonomics Podcast episode (and talks about how immediate feedback interventions– like the use of anesthesia in surgery–catch on much more quickly than delayed-feedback ones — like antisepsis).

I wish I could end this post with some brilliant solution to this lack of feedback, but for now, I only have the observation. I suspect the problem is less prevalent in veterinary care both because of less emphasis on privacy (no HIPAA for pets), and because of less complexity and specialization in the systems of care. I’d like to think that Atul Gawande is right when he says in this podcast that we’re in the MS DOS phase of medical information systems, and that at some point in the future, systems will be integrated enough to provide medical providers with the feedback they need.

As for me, other than being determined to climb bigger mountains than my current fitness level would indicate, I am doing well. My incisions are all healed and my various tubing has all been removed. Other than follow-up with occasional ultrasound exams, I’m done with this whole process of complication and repair.

I’ll end with a few more pictures from my hike last week on Mt. Cube (scroll back up to the beginning of this post to see the panoramic view from the mountaintop).

While I can only speculate about the causes of the surgical error in which I was a participant but not a witness, I do know a few things about errors in surgery. One of the most confounding things about surgical errors (besides figuring out how to recover from them) is why they happen at all. Why, when I do the same procedure the same way every time, does this one time result in a complication?

Previous authors have spent a lot of time thinking about error taxonomy in surgery, anesthesia, and elsewhere, and while the answers aren’t always simple, they can be categorized. (The types of error that can happen in surgery are summarized nicely in this article by Cuschieri.)

Errors in execution or perception

Many surgical (and anesthetic) errors stem from differences in individual patient anatomy or physiology that reveal vulnerabilities in an existing technique or protocol—a technique or protocol that is adequate in most but not all circumstances. These can be some of the most difficult errors to identify and understand. An example might be a veterinarian who tightens the ligature around every cat’s uterine body and vessels using one pound of pressure on the first and second throw of her ligature knot. In most cats, this will be sufficient, but in a few (probably in-heat) cats, it likely will not, and those cats may bleed from the ligated vessels and perhaps require re-operation and re-ligation.

This inadequate ligature tension is an error in technique that in most cats causes no harm.  In each cat, there is the possibility that it will be adequate, or not. The solution to this error would be to tie all uterine body ligatures with greater tension (say, 2 pounds of tension), or to use a cue from the patient to determine when ligature tension is adequate in that patient (such as observing tissue blanching under the ligature).

Other errors in execution in surgery may be due to lapsed attention or to a mistaken perception. Attention lapses are more plausible than one might think: proficient surgeons use habitual motions and automatic processing, rather than the more deliberate and slow techniques of the inexperienced surgeon (again, see the article by Cuschieri). This automaticity allows for efficient and minimally traumatic surgical performance and is essential to practice, but it may also allow for inattention. Often if a problem arises, the surgeon will perceive an inconsistency between her mental picture of the surgery and real life, which will focus her attention and allow her to adapt to the new surgical circumstances. If not, the problem may go unnoticed and result in a complication.

A mistaken perception may go hand in hand with automaticity or with inexperience. Anatomy varies, and the surgeon may not see what she expects to see where she expects to see it. A misidentified organ, vessel, duct, or tissue can easily be severed or ligated.

Latent (system) errors

Another type of error that happens in practice may be a systematic error, a problem in the organization and allocation of time and resources. This latent error is a gap that is present at all times, but may go unnoticed and uncorrected, especially if no harm comes from it.

Professor James Reason makes the analogy of Swiss cheese, in which each slice of cheese is a layer of defense, and it is only in certain unlucky circumstances when the holes in these layers align to allow an accident or injury to occur.

In veterinary practice, an example of a latent error might be limited or infrequent observation of patients recovering from anesthesia. In perhaps 99.8% of patients, there will be no problem if they are observed only intermittently during the period after they have begun to rouse from anesthesia but before they are fully ambulatory. But in a few patients, that gap, that latent error, will be the hole in the system through which they fall. (The post-op period is the highest risk period in small animal anesthesia – perhaps because of the inherent danger of this time period or perhaps because of this common latent error of limited observation post-operatively.)

In many practices, especially with good luck and a low volume of surgeries, this particular organizational practice may not result in a patient death for many years. If and when one finally does occur, it seems to happen out of the blue, but is really a combination of this individual patient’s physiology combined with the latent error in observation, perhaps also combined with other transient or latent circumstances.

If you are interested in learning more about human error and error taxonomy, I would recommend any of the books or articles by James Reason. For some excellent and thoughtful essays on complications and learning in surgery, read Atul Gawande, especially his books Complications, Better, and the Checklist Manifesto.

As for me, I’m doing well. Yesterday I went snowboarding and got to enjoy the remnants of snow and a blue-sky sunny day. An extra wrap of spandex around my waist (a maternity product borrowed from my sister) kept the tube and pouch snug and in place, so no wardrobe malfunctions during my outdoor fun.

Frequently asked question: I have been asked by several people whether I still need to pee “the old fashioned way”after my recent nephrostomy tube placement.  Yes! My left kidney drains into thenephrostomy bag (my newfangled bladder), while the right kidney still has a good ureter and drains into my old fashioned bladder. That means that I have to pee twice every time I go to the bathroom since I have two bladders to empty. Downside: bathroom visits take twice as long. Upside: I only have to go to the bathroom half as often.

On the same day my article on coping with complications was published online, I met with my doctor.  I was recovering slowly from my hysterectomy and eight weeks along, my belly was still sensitive and swollen.

A week later, I was in the hospital, one tube draining my belly, another in my left kidney. My left ureter had been damaged during my hysterectomy, and urine had been leaking into my abdomen. The first drain would remain in place for 10 days to relieve the abdominal pressure and drain the urine that had accumulated in my abdomen, while the drain in my kidney, a nephrostomy tube, would send the urine from my left kidney into a bag I would carry with me for months like an external, second bladder.

It will be two months before the urologists surgically reimplant my left ureter into my bladder, before they dissect through the scar tissue and reveal the injury. Until then, we won’t know the nature of the injury or what could have been done to prevent it.

In veterinary surgery, ureter damage during ovariohysterectomy is rare enough that this complication isn’t reported in most studies of spay neuter complications (one summary of the literature on spay neuter complications mentions only 3 case reports ). Ureter damage during spay is like the monster under the bed: it is the star of horror stories that happened to someone else, never actually seen or experienced by you or anyone you know but still terrifying and perhaps possible.

The strange, compacted, biped anatomy of humans, though, places the ureters in the surgical field and at risk during hysterectomy. Ureter damage happens in about 0.5-2% of hysterectomies in women and can be the result of ligation, kinking by suture, transection/ avulsion, partial transection, crush, or devascularization. Exactly which of these happened may become clear during my ureter repair this spring, but until then is only guesswork.

So my doctor is left in that unenviable circumstance of having an unresolved complication of indeterminate cause. The long delay between my surgery and my return with symptoms would make the details of my surgery itself difficult or impossible to recall. The unknown error (and known result) leaves her hanging, no guidance for improvement, no specific oversight or action to correct in her next surgery, or the surgery after that. Only worry, and caution. I recognize the feeling, and the fear.

At my initial surgery, the humor in being a spay neuter veterinarian undergoing a hysterectomy wasn’t lost on me, nor now is the unfortunate coincidence of writing about surgical complications and then experiencing one. I appreciate the perspective and can’t help but use it as an opportunity to keep exploring the experience from within the domain of human factors and ergonomics. In addition to considering surgical complications, my experiences of the last few weeks and months have led me to think about topics as diverse as error taxonomy and latent error to organizational leadership. Between now and my repair, I plan to explore these topics here. Stay tuned!

As I wait for my repair, I am fine and happy, and feeling physically better than I have in months, despite the nephrostomy tube and external “bladder.” This morning I went for a 9 mile ice skate at Lake Morey, and it felt good to be out on the ice and moving on such a beautiful day.

So much ice and blue sky!

Sign on the ice by the pond hockey area. No, I didn’t. But I can pee standing up now…