Undifferentiated Flank Pain: The Case for 2 CT Scans
During our GU board review this week, I got to thinking about kidney stones and the frustrating stepwise approach to the patient with equivocal flank pain and a working diagnosis of nephrolithiasis. Most cases are easy since renal colic is common and presents so characteristically, but there are some patients with equivocal presentations. These patients with mushy histories, maybe some mild flank pain, maybe not, possible dysuria, and subjective fevers present diagnostic challenges. Is this an episode of renal colic, renal colic with obstruction and superinfection, pyelonephritis, or something else in the back or belly? This is when the first groan happens because the realization is the that the patient is possibly going to need 2 CT scans: one without IV contrast and then another one with. But maybe that’s not necessary because pyelo can be seen on a non-contrasted CT right? Something about fat stranding? Maybe we can just get away with the single scan, and if we don’t see the stone, we’ll find something to hang a weak diagnosis of pyelo on.
A quick review of nephrolithiasis. The most general approach to these patients is pain control followed by urinalysis. As was mentioned at conference, pain control is best achieved with combination therapy using an NSAID such as ketorolac and narcotic such as morphine (PMID: 16953530). If the clinical presentation is correct and urinalysis shows blood, imaging is not necessary. Here’s the first problem: the “classic” finding of hematuria can be absent up to 15% of the time (PMID: 7747369).
In cases which the diagnosis is not as clear, non-contrasted CT scan of the abdomen is indicated and proves to be a very good test for nearly all renal stones except for those secondary to HIV protease inhibitors such as indinavir (PMID: 9230000). So that’s great because unless we are dealing with that specific situation, the non-contrasted CT scan should certainly find the stone. But invariably that sometimes doesn’t happen, which brings up the second problem: what now?
At this point the arrow starts to point further down the differential and a diagnosis of pyelonephritis comes to mind given the patient’s bloody urine, positive leukocyte esterase, and mild flank tenderness. While pyelo is a clinical diagnosis supported by a characteristic urine, often muddled histories and unconvincing exams can leave the diagnosis in doubt. Non-contrasted CT scans can show an enlarged kidney or perinephric fat stranding indicative of a pyelo, but they can also be normal in the setting of pyelo. Thus in the patient with an unclear diagnosis based on urine and history and non-diagnostic CT scan, a contrasted CT scan should be pursued as the next step(PMID: 16937102, 15486235). Besides giving a much better image of the kidney in cases where the diagnosis is in doubt, a contrasted scan will highlight vascular diseases or renal infarcts whose presentations mimic that of renal colic and would not be seen on a non-contrasted scan. These are rare entities but can be missed easily. Bottom line: slog through getting 2 CT scans when kidney stones are not seen on the initial non-contrasted CT and the clinical picture is not a slam dunk for pyelo - the correct imaging for pyelo should be with IV contrast.
Posted
on Friday, October 19th, 2007 at 9:22 am by Sohan. Filed under
GU, Infectious Disease.
You may post a comment.
why wouldn’t a ct scan pick up indinivir stones????? also i would add that in patients with prior stones and a stone history a bedside renal ultrasound can be performed to rule out hydro, obviating an ct scan, assuming pain was controlled and no fever.
also… there are those rare renal infarct pts with a neg ct minus, so push for a + if you suspect, esp with a high ldh.
Comment by bcs on December 10th, 2007 at 8:34 pm
First, I am not sure why a CT scan would ever be needed to diagnose pyelo. Even with a poor historian a physical exam and a simple UA will reveal pyelo. If your concern is a renal abscess then perhaps you have a case.
Second, even if you are sure clinically that ureterolithiasis is the cause of flank pain there are times when the CT scan is read as negative. Keep in mind that reading a CT scan is also done by another human who is prone to error. At some point you have be comfortable with a clinical diagnosis. CT scan is a great tool, but we are the diagnosticians– not the machine.
Most of these cases can be diagnosed clinically. In most ED’s you are hard pressed to manage resources and get one CT scan. Getting two CT scans is a bit of a luxury, further exposure to radiation and with what type of yield?
If you are seeking rare causes of flank pain then these work ups should be tailored to those in the proper age groups with appropriate risk factors.
Comment by HLC on March 7th, 2008 at 7:55 am