Abdominal Exam: are we just wasting our time?

To start with a difficult admission: I have a fairly long history of doing a poor abdominal exam, despite being very interested in this bedrock of the physical examination. In fact, I love the physical exam, in theory; but I have recently found that, in practice, I have been sorely lacking in both technique and the willingness to perform up to my knowledge.
How this came about:
I was recently working in a small emergency room with a group of non-emergency-trained physicians. Old school guys. A bit behind on the newest information, but overall not bad physicians. I went in to see a woman complaining of abdominal pain, got an appropriate history, and then performed my abdominal exam. Maybe a little questionable, but overall a benign exam. I went to chat with one of the docs about her and he went in to do his abdominal exam. Acute abdomen. CT scan showed florid diverticulitis. She was admitted to the hospital for IV antibiotics.
The very next day, a middle-aged woman came to the ED complaining of, what else but lower abdominal pain. I decided to test myself. After taking her history - relatively unimpressive (resolving abdominal pain, currently about 2/10 and more "achy" than really "painful") - I began with my "old" (I now refer to as "crap") abdominal exam. Benign in the truest sense of the word. Then I did a thorough exam. Acute abdomen with rebound tenderness localized to the right, lower quadrant and a strange variant of Rovsing's sign (palp in the LLQ made both LLQ and RLQ hurt). Suddenly suspicious of significant pathology, I ordered a CT and... acute appendicitis. She was wheeled off to surgery.
The real lessons I learned were to push hard - it may hurt more, but it also tells you more, and if you must, give the patient morphine first (it actually seems to bring out pathology rather than cover it up) - and do a real, honest-to-God rebound tenderness test - jerk your hand off the patient's belly (it really does make a difference). Furthermore, when evaluating for rebound tenderness, try in several locations. The lady with the appendicitis only reacted when I tested her RLQ. And the heel jar test does NOT necessarily correlate with peritoneal irritation. Both ladies had a negative heel jar but positive rebound tenderness.
It's too late to get into it right now, but I want to incorporate this with a discussion on the use of CT scans. The SMART EM podcast has a fantastic discussion of the risks of CT scans (I highly recommend it) that would go well with a discussion of the utility of an excellent physical exam. There was also a recent article concerning the use of serial abdominal exams rather than CT in the pediatric population for diagnosing acute appendicitis. Not to mention some excellent scoring systems (the one I am most familiar with is the Alvarado Score). I'll try to get back to this later.

MONA or just ONA?

The age old acronym of "MONA" used in the management of acute coronary syndrome has recently come under some criticism after a study published in 2005 in the American Heart Journal noted that patients treated with morphine had a higher mortality rate than those who did not recieve morphine. Two subsequent letters to the editor differed in their reception of this information (as good science would have it). So, what is the gist?

The study itself - "Association of intravenous morphine use and outcomes in acute coronary syndromes: Results from the CRUSADE Quality Improvement Initiative" - was a non-randomized, retrospective, observational trial. It enrolled patients presenting with non-ST-segment elevation acute coronary syndromes and ultimately noted an increased odds ratio of death of about 1.5. Considering its limitations, the study was well done and run through several re-assessments, including risk adjustment (i.e. comparing the sickest patients given morphine to the sickest patients not recieving morphine) and matching on propensity score for treatment (most basically, an attempt to minimize the influence of confounding variables). Ultimately, there needs to be a randomized control trial on the issue, but this is still concerning.

The letters to the editor (December 2005; American Heart Journal):

#1: Observation that morphine supresses ("paralyzes") cortisol production in the opioid-naive, potentially dropping levels as much as 75% or more within 3 hours. Occasional Addisonian crises. Notes that some reports suggest improved survival in MI patients treated with corticosterooids (I am personally not familiar with this information), possibly due to a replacement effect rather than the oft cited anti-inflammatory hypothesis.

#2: Although the original article cites a source that suggests morphine increases infarct size in animal models, this author notes that in his own lab, morphine reduces infarct size. He does note, however, that the effect is largely dose-dependent and time-dependent, showing no effect at higher doses and being effective only when given before and ischemic event and just before reperfusion, losing their cardioprotective effect when administered after reperfusion. How this compares to the other study that suggests that morphine not only does not work but is harmful, I haven't figured out yet. I'll update when I do.

Medical charting

It is customary, at least in the Emergency Department, to have selective hearing of sorts. A patient says, "I have low back pain." The ER doc asks the requisite, "do you have any problems with urinary incontinence or do you ever poop on yourself?" The patient replies, "well, I have peed on myself once or twice, but not in the last few days." The physician writes down, no urinary or fecal incontinence.

This is not unreasonable, as an affirmative response to either of the questions calls for a lengthy and expensive work up, steroid administration, and possible neurosurgery or orthopedic consultation, and this patient's back pain does not seem to be related to their urinary incontinence. However, a falsehood (of omission) has now been written on this patient's chart. And this is done dozens of times, by dozens of doctors every day. For practical purposes it probably makes little to no difference. Patients receive adequate emergency care regardless of whether or not their 2/6 systolic murmur was recorded on their chart; but, if ever a chart review for research is conducted, inaccurate information is obtained. If a physician tries to find correlated physical findings he may draw an empty net, as many of the findings may have been omitted.

But I can't blame the physicians. If given the choice between omitting information on the chart and saving myself from a possible lawsuit or writing down every tiny detail for future posterity and possibly having the shoes sued off of my feet, I would omit. Because every detail on a chart must be explained. If a chart has "+ urinary incontinence" in a patient with low back pain, a physician must have a good, a very good reason for not performing an MRI. It doesn't much matter if they've had urinary incontinence for 30 years. A lawyer will pin them to the ground. And until this changes, physicians will lie.