Friday, July 29, 2011

Is Robotic Surgery All Hype and No Substance

I am unable to contain myself for another minute. Let’s talk about robotic surgery.

This is from How Stuff Works, a Discovery Company, but it could have been written by the makers of the surgical robot:

Most surgeries require nearly a dozen people in the room. As with all automation, surgical robots will eventually eliminate the need for some personnel.”

This is another example of a “straw man.” [Establish a false premise and defeat it with your argument.] Just about all operations require five people—surgeon, assistant, scrub nurse, circulating nurse, anesthesiologist. Some complex procedures might require another assistant. A dozen people might be needed for separation of Siamese twins or bilateral leg transplants.

There are many theoretical advantages of robotic surgery over conventional laparoscopy such as elimination of hand tremors, availability of more flexible instruments, more precise movements, easier learning and execution of intra-corporeal suturing techniques and possibly others. Disadvantages include the costs associated with robotic technology with the robotic itself selling for $1-2 million, yearly service contracts of >$300,000 and the high cost of the specialized instruments. Also, robotic surgery takes much longer than conventional surgery.

The problem with robotic surgery is that it has never been shown to improve patient outcomes for any procedure. Let’s look at the literature. The review articles cited below are almost exclusively based on non-randomized studies.

Laparoscopic Cholecystectomy. A review by the noted Cochrane Group involving five studies and 453 patients showed no differences in any outcome measure when comparing robotic surgery to conventional laparoscopic surgery.

Esophageal Reflux Disease. A review of 11 papers comparing standard laparoscopic anti-reflux surgery to Robot-Assisted Laparoscopic Fundoplication [acronym “RALF”] in 533 patients showed no differences in peri-operative complication rates or length of hospital stay. The robotic procedure took significantly longer. Another recent paper demonstrates the lack of quality research on this topic.

Colorectal Surgery. A review of 17 studies, one of which was randomized and controlled, showed no difference in the rates of complications and cancer outcomes. Robotic procedures took longer an cost more than conventional laparoscopic colon surgery. Despite the results, the authors felt that “Robotic colorectal surgery is a promising field and may provide a powerful additional tool…”

Gynecologic Surgery. A review of 22 non-randomized studies found that robotic surgery resulted in less blood loss [statistically significant but not clinically significant differences] and shorter hospital stays but no differences in overall complication rates when compared to conventional laparoscopic or open surgery. The authors commented that the methods used in the papers reviewed were poor and better studies are needed before concluding that robotic surgery offered any true advantages.

Abdominal Surgery. A paper from 2010 looked at 31 studies of nine different abdominal operations [robotic vs. conventional laparoscopic], 6 of which were randomized, controlled trials [RCTs]. The total number of patients included in all the studies was 2166. The number of patients who were participants in RCTs was 230. No RCT involved more than 50 patients. Not surprising was that the results were mixed with robotic surgery offering no clear advantage. These authors also called for larger and better designed studies.

Prostate Cancer. To date, there are no good RCTs comparing robotic to open or standard laparoscopic prostatectomy. This quote from a recent review of the literature on prostate cancer surgery says it all:

Robotic prostatectomy is definitely here to stay and although a randomized, controlled trial comparing the open to robotic techniques would be ideal, it is clear that this is unlikely to occur.”

UPDATE [9/17/2011]
Robotic hysterectomy outcomes are no better than standard procedures reports a new study. It also suggests the procedures are driven by marketing.

UPDATE [12/29/2011]
Add gastrectomy to the list of procedures that do not show better outcomes when done robotically. 

UPDATE [1/3/2012]
ENT surgeon, @FauquierENT, dismisses robotic surgery for sleep apnea as "way overkill akin to using a $50,000 sniper rifle to kill an ant on the wall."

The issue may already be settled. According to the New York Times, patients are voting with their feet, preferring to have robotic surgery when it is available. This appears to be true in Wisconsin as well. A recent paper reveals that when hospital purchase a robot, their volume of prostate cancer surgery doubles.

Questions. Can anything be done about this? Should Medicare and private insurance companies pay for expensive, unproven treatments?

Wednesday, July 27, 2011

Bad News For MDs. Researchers Work on Vocabulary for Chronic Pain

Just released on Science Daily is the news that psychiatrists in Buffalo are working on a project to enable chronic pain sufferers to better describe what they are feeling. The docs are using ontology, the branch of metaphysics that studies the nature of existence or being as such. Metaphysics is sometimes called philosophy, especially in its more abstruse branches. [Definitions are from]

Here is a quote from the Science Daily article:

"The philosophical definition of ontology is the study of things that exist and how they relate to each other," says [Werner] Ceusters, who also is director of the Ontology Research Group of UB's New York State Center of Excellence in Bioinformatics and Life Sciences. "I am a person and you are a person so we share something. Suppose I drop dead. What lies on the floor? Is that still a person? If it is no longer a person, is it still the very same thing that was sitting here as a person but now is a corpse?"

If you can explain that quote to me, I would certainly appreciate it.

Someone must understand it because the group has received an NIH grant of $793,571 to study the subject.

You may have no idea how many chronic pain sufferers there are. I do. For fibromyalgia [also known as chronic widespread pain syndrome] alone, 17% of a population in a published study from England had it. In my blog on the paper, I pointed out that at the rate people were developing the syndrome, 50% will have it by the year 2033.

Honestly, I do not see standardizing a vocabulary for pain based on ontology leading to anything good.

PowerPoint Continues to Obfuscate

Despite years of evidence showing that PowerPoint can make you stupid, its use continues unabated. Here are some selected slides from a recent military briefing. 

Rather busy slide
Way too many words
It's all clear now
Time line a bit too cluttered
TMI (Too Much Information)

If you dare, you can look at the complete set of briefing slides here.

Monday, July 25, 2011

The “Dumbing Down” of the US Citizenry Thanks to Product Liability Litigation

Everyone has heard about those silly product warning labels that are made necessary by constant litigation. For example, here is one from a drain cleaner.

It says, “If you cannot read the directions, don’t use this product.” But if you can’t read, how do you know what it says?

Here is another one for you. We bought a pool toy that consists of several streamers that stand up in the pool. There is a weighted bottom with a buoyant top. (See photo below.) Children can swim around them under water or dive down and grab them. Sounds like fun, right? 

Of course it’s fun. But the manufacturer, Speedo, is taking no chances. On the bottom of each toy is this warning:

So, wait a minute. I shouldn’t throw something with a weight on it to a drowning person?

I happen to have a couple of extra bricks in my backyard. I am thinking all bricks should carry labels stating, “This is not a lifesaving device.”

Friday, July 22, 2011

Robotic Single-Port Cholecystectomy Fails to Convince

I have resisted the urge to blog again about single-port [a.k.a. single-incision] laparoscopic cholecystectomy because I was afraid I would start swearing a lot, but I can contain myself no longer. The stakes have escalated with the appearance of a publish-ahead-of-print paper on robotic single-port laparoscopic cholecystectomy in Archives of Surgery.

One of the co-authors of this paper is the chief medical advisor for the company that makes the robotic surgery device and the other is a consultant. The paper is a pilot study of the feasibility of using the robot to perform single-port laparoscopic cholecystectomy. It attempts to retrospectively compare 10 patients done robotically to 10 patients who had standard, four-port laparoscopic cholecystectomies. Patients with acute cholecystitis were excluded from both groups.

One robotic case had to be converted first to a four-port laparoscopic approach and then to an open cholecystectomy. This patient was excluded from the analysis of the results, a flagrant violation of the “intention to treat” principle which is that all patients should remain in whatever group they originally were assigned to.

The robotic cases [minus, of course, the case that was converted] averaged 105 minutes in duration which was not significantly different than the four-port cases, which averaged 106 minutes. The authors concluded that this showed that the duration of robotic single-port laparoscopic cholecystectomy was equivalent to four-port laparoscopic cholecystectomy. Sounds great, right? Not really. This is a nice example of what is known in the statistics business as a “straw man.” A straw man is defined as establishing a control that is not representative of real life and then comparing the experimental group [favorably] to it. Most four-port laparoscopic cholecystectomies for non-inflamed gallbladders can be done in well under 60 minutes. For example, another recent paper comparing non-robotic single-port laparoscopic cholecystectomy to standard, four-port laparoscopic cholecystectomy noted mean case durations of 88 and 45 minutes respectively.

Of the nine analyzed robotic patients, two suffered urinary retention, one of whom had to be discharged with a catheter. This is a very rare complication of four-port surgery.

The authors state, “This approach [robotic single-port] appears to be safe, even in difficult cases with inflammation, and has a high degree of satisfaction with the patients.” It is difficult to see how they arrived at that conclusion since patients with inflammation were specifically excluded from the study. Patient satisfaction was only briefly discussed and not compared to the standard surgery group.

Not mentioned in the paper is the cost of the robot which is at least $1.3 million plus yearly maintenance fees of hundreds of thousands of dollars. All of this is being advocated without any evidence that the robot is safer or more efficacious for any type of surgery, let alone standard laparoscopic cholecystectomy. And it is being promoted to avoid three 5 mm incisions which in most patients are not painful and barely visible if at all. 

All I can say is "Klaatu barada nikto," which some have interpreted to mean “This escalation is unnecessary.”

Thursday, July 21, 2011

Choosing a Medical Specialty Is Difficult

A blog that was posted two months ago just came to my attention. On the KevinMD website, a physician wrote about the trials and tribulations of emergency medicine. The piece was a description of how the author chose her specialty. Looking for balance between work and family, the author picked EM because she “…thought the shift work in EM would allow me to have the best of both worlds.” After she had children, she discovered that juggling career and family responsibilities caused significant emotional and physical stress. It turns out that emergency rooms are open 24 hours per day and seven days per week. This necessitates staffing even in the wee hours of the morning. This “…did not fit well with family life.”

She goes on the say “Is the toll that working odd hours [takes] worth it? Is the exhaustion EM physicians experience jeopardizing patient safety?”

The author asks, “What are possible solutions to this problem since the highest need for EM clinical coverage is in the evenings and nights when office-based doctors have already gone home?” I can’t think of many.

There are two major issues.

1. Work hours. Hold on a minute, I thought the fuss was about docs working inordinately long hours, like 30 at a time. You mean to tell me that even a 12 hour shift [the longest stretch that any ED MD has to work] causes exhaustion to the point of jeopardizing patient safety? How can this be? Don’t let the ACGME find out or everyone will be working 4-hour shifts. Can we close emergency rooms at, say, 5 pm? I’m not sure the public will buy into that since they are already up in arms about all things related to medical care. And who would order the CT scans?

2. Family considerations. With a few exceptions such as dermatology, radiation medicine and rehabilitation medicine, all medical specialties interfere with family life. Either practice one of the previously mentioned “9 to 5” specialties or consider a career change.

Can you think of any other solutions to this problem?

Wednesday, July 20, 2011

Twitter MDs Can Change the World. Not so much.

On Tuesday of this week in response to somewhat of a groundswell of enthusiasm about physicians needing to have a presence on social media, I posted a blog about MDs relating to potential patients on Twitter. I received a thoughtful comment agreeing with me that MDs probably will not be able to use Twitter to market themselves or engage in discussions with patients. The commenter went on to say that Twitter might be a vehicle for communicating and educating, maybe even altering public opinion. She said has only 100 followers but is being followed by someone with 1000 followers. Therefore a tweet that she makes could conceivably be retweeted many times.

I don’t think so.

Before I read that comment, I was going to call this post “How Twitter Makes Me Feel Insignificant.” Here’s why.

According to PeerIndex and Klout [two social media rating services], I am well above the mean for influence on Twitter. I thought, “Well, that’s pretty good for someone who has only been on Twitter for a year.” Then reality reared its ugly head.

The population of the US is estimated at 312 million people. Only 42 million or so US residents use Twitter. I have about 570 followers. This represents about 0.00018% [560/312M] of the US population. Of my 570 followers, at least 25-30% are commercial organizations or individuals. I’m not sure how religiously they are following me. In addition, many of my followers are following well over 100 people. I am following about 70 people and can’t keep up with their tweets. I realize some people may be using lists to narrow their focus. But I'm listed by only 48 people. All this leads me to doubt that many people are holding their breath waiting for my next witticism.

Only about 20 people regularly respond to my tweets with replies or retweets and they don’t all respond to the same material. If I’m lucky, I might get five replies/retweets for a single tweet of mine. Amazingly, this makes me influential. I’m not a “rocket surgeon” so I don’t exactly understand how this could be.

Most doctors on Twitter have rather modest numbers of followers.

I just don’t see how we will be changing the world with tweets. On the other hand, you might be satisfied if you can enlighten one person about an important issue. Maybe that's good enough. [Yikes! A ray of hope from a confirmed skeptic.]

Tuesday, July 19, 2011

Physicians and Twitter Reality Check

Some [here and here] are saying that physicians must have a presence on social media. I was thinking about this and decided to see what is going on with Twitter. Here are some numbers to ponder.

Twitter has 106 million users. Over 60% or 63 million are outside the US. Some 47% or 50M Twitter users are under the age of 35.

The population of the US is now about 312 million.

Let’s say you live in a city of 100,000 people and the same percentage of Twitter users in the country used Twitter in your area. [106M x 40% = 43M/312M = 14% or 14,000 people.] Now deduct the % under the age of 35 as they are unlikely to need a doctor. That gives you 14,000 x 53% [the % over age 35] which = 7420. You would have to assume that most of those 7420 people would not be following you on Twitter. Why? According to a report published in 2010, 60% of Twitter users follow fewer than 10 people. Assuming you are not being followed by that group, you now have 6420 x 40% or 2968. Since 74% of Twitter users have fewer than 10 followers, what is the likelihood that even a few of them would be following you? I suppose you could give your Twitter handle to all your patients. Do you really want to do that? What if a patient tweets something significant about a change in his symptoms and you don’t respond promptly or misinterpret what he said and give the wrong advice?

To look at it another way, over 92,000 people in your city will not even be potential viewers of your tweets.

If you are looking for a way to waste some time, get a few laughs, commiserate with like-minded physicians and occasionally learn something, Twitter might work for you. If you think of it as a way to market yourself or communicate with patients, I suspect you will be disappointed.