Diagnosing kidney stones is relatively easy, and the treatment is relatively straightforward. A modicum of rather serious problems, such as renal infarction, retroperitoneal bleeding, renal abscess, and ruptured abdominal aortic aneurysm or an aortic dissection, can mimic a kidney stone, so let the clinician beware. Aortic catastrophes can even be accompanied by the hematuria that is common, but not omnipresent, with a kidney stone.
The past few columns have highlighted several issues, including that intravenous fluids do not hasten the passage of a kidney stone, nor will hydration likely worsen the pain. Likewise, an abdominal radiograph is a useless endeavor that only wastes time and misleads the clinician, while a nonenhanced spiral CT scan quickly settles whether a kidney stone is present, assesses the possibility of obstruction, and adroitly detects a plethora of other medical problems with similar presentations.
I advocate for almost universal use of a CT scan to come to a quick clinical conclusion. Many clinicians, however, will use their clinical acumen to diagnose a kidney stone in selected patients. Few urologists have such bravado. If the patient is 34 years old, and presents with a chief complaint of “I am passing another stone,” and the history and physical are otherwise consistent, some will eschew routine CT scanning, at least in the ED. This approach has support in the literature and in clinical practice. I prefer to call the question, get the answer, and proceed with aggressive therapy while being off to see another patient.
Once the prescient and sagacious clinician has nailed down the diagnosis, providing analgesia becomes a priority. Certainly we should offer the most reasonable, aggressive, and logical approach. It's not rocket science that a kidney stone is no time to attempt to save money by being parsimonious with morphine, Dilaudid, or ketorolac. This month's column tackles another related issue: infection in the presence of a calculus.
Midstream Urine Culture and Sensitivity Test is a Poor Predictor of Infected Urine Proximal to the Obstructing Urethral Stone or Infected Stones: A Prospective Clinical Study Mariappan P J Urol 2004;171:2142
This recent article addresses an issue that receives little attention in the emergency medicine literature. It attempts to determine whether a midstream clean-catch urine culture is an adequate test to define the presence or absence of a urinary tract infection in patients who have an obstructing kidney stone. The authors reiterate the common axiom that an obstructed ureter with stagnant urine is a potential source for serious infection. In general, the definition of the UTI is based on an examination of the urine for bacteriuria, leukocytes, and eventually a urine culture. That usually means obtaining urine from the bladder, either with a spontaneously voided specimen or a catheter.
In urologic circles, prior to elective endoscopic manipulation of uncomplicated stones, it is desirable to have a negative urine culture. In addition, procedures performed to remove stones (ureterorenoscopy [URS] or lithotripsy) can themselves introduce systemic infection. Prophylactic antibiotic coverage is the norm prior to urinary tract manipulation, even when dealing with a sterile milieu. When possible, urologists avoid elective stone manipulation, surgically or with ultrasound, until they confirm a sterile culture. In some cases, the stone itself is infected, and fragmentation of the calculus can release bacteria into the urine and then systemically. In the ED, however, decisions about infection in the presence of a stone can be problematic, yet important clinical interventions must be made with limited information.
The authors indicate that in their practice, patients undergoing elective ureteroscopic procedures all will be subjected to a midstream urine culture and sensitivity a week prior to the procedure. If an infection is proven, antibiotics are administered prior to any surgical intervention. Another sterile urine is required prior to rescheduling the surgery. Even if the patient has a sterile urine, prophylactic antibiotics are used during the procedure.
Patients excluded from the trial were those at high risk for infection, including those with renal failure, diabetes, immunocompromise, or previous manipulations or urologic procedures. Essentially, these were otherwise healthy patients who had a stone, and represented a population unlikely to become septic from either the stone or the procedure. Despite one's best intentions, an obstructed ureter and a proximal urine infection often add up to sepsis in rapid fashion. These authors attempted to determine whether a midstream urine culture and sensitivity — urine that is distal to an obstruction — could reliably identify infected urine proximal to the stone.
In this prospective clinical study, 73 patients undergoing URS or lithotripsy were first subjected to a voided midstream urine collection for culture and sensitivity analysis. Prophylactic antibiotics (ampicillin/sulbactam) were given prior to operative intervention to remove the stone. Using a guide wire, a catheter was passed proximal to the obstructing stone to obtain a proximal urine sample for culture and sensitivity. The results of this culture were compared with the bacteriology of the urine obtained by a standard pre-op midstream catch. Lithotripsy was then performed. The stone fragments obtained also were sent for culture, attempting to identify whether it was an infected stone. Patients then were followed closely for signs of bacteremia or sepsis.
Patients in this study ranged from 17 to 49, with a mean age of 37. This correlates with previous data confirming that kidney stones are not usually a disease of the very young or the very old but primarily one of young and middle-aged patients.
About 28 percent of the patients had a positive urine culture when the distal urine was obtained in the standard midstream fashion. At baseline, many patients had infected urine already. Three times as many patients, about 60 percent, demonstrated a urine infection when the urine proximal to the stone was cultured. About a third of the stones cultured were positive for infection, a so-called infected stone. Interestingly, a stone culture and the proximal urine culture were positive simultaneously in only 23 percent of cases.
Clean-catch and stone cultures were simultaneously positive in 11 percent of cases. Most importantly, the midstream and proximal urine cultures were both positive in only 16 percent of cases. Overall, the preoperative midstream culture had a sensitivity of only 30 percent to detect a positive culture in urine proximal to the obstructing stone. The conclusion was that an analysis of urine distal to an obstructing stone is a poor predictor of the presence or absence of infection proximal to the obstructing calculus. The positive and negative predictive values of the distal urine culture and sensitivity were quite poor.
Clinical bacteremia was noted within 24 hours of the procedure in almost 20 percent of patients. Half of these patients became quite ill with multisystem failure and hypotension and needed inotropic support. Of the patients who had clinical bacteremia following the procedure, none had a positive midstream urine culture preoperatively. It appears that the procedure precipitated the infection by manipulating an infected stone or disturbing the proximal urine.
The organisms obtained were predominantly E. coli, but Proteus, Acinetobacter, Pseudomonas, and Klebsiella also were recovered.
Comment: I was quite flummoxed by the results of this study. It indicated an infection rate far greater than I would have assumed or have found in my 34 years of clinical experience. To my recall, very few patients who present with renal colic have coexisting infection. Perhaps that's because we usually see otherwise healthy uncomplicated patients with simple renal colic. The patients in this study were a different breed. They appeared to have long-standing obstructing stone, and had stagnant urine behind the stone. This study is from Malaysia so perhaps sterile techniques or stone bacteriology are different, but there was an amazingly high incidence of co-existing urinary tract infection and kidney stones.
These data can be applicable to some ED patients, however. It's important to know if a patient has complete obstruction from the calculus, a prerequisite for developing an infection in stagnant urine. A minor degree of partial obstruction manifested as hydroureter is a common finding on CT scan in uncomplicated stones. Usually this does not change disposition/therapy in the ED. This article would suggest that it is essentially a worthless endeavor to obtain a urine culture and sensitivity from a patient with a kidney stone who has complete obstruction. The culture results of distal (bladder) urine simply do not rule in or out infection proximal to an obstructing stone. Most importantly, a negative culture can be quite misleading. The incidence of bacteremia following stone removal in this study also was quite impressive, about 20 percent. The value of obtaining urine proximal to the obstruction was stressed, but this technique is out of the realm of emergency medicine.
While it is clear that an infection proximal to a kidney stone is a ticking time bomb, I know of no standard of care in emergency medicine that mandates even a routine voided urine culture in all patients with seemingly uncomplicated ureterolithiasis. In fact, most emergency physicians never send a urine culture or think of infection in patients with renal colic. A standard urinalysis is routine, but a few white blood cells or trace readings on the dipstick for leukocyte esterase or nitrites rarely get your attention or prompt additional interventions. A few white blood cells in the urine is consistent with stone-induced inflammation. For most patients, we are generally happy if we make the diagnosis and are able to find a urologist to participate in the care. Many patients with stones are sent home, even with partial obstruction secondary to the calculus. I also know of no standard that mandates routine antibiotic prophylactics in patients who are discharged with stones so the results of this study perplex me.
I have not seen other reports in the emergency medicine literature studying routine urine culture in patients with kidney stones, and even the urology literature lacks studies comparing urine cultures proximal and distal to an obstructing stone.
This is a nifty study. The authors claim that this is also a first in the urology literature. It's probably more applicable to urologists performing surgical interventions on patients with long-standing complete ureteral obstruction, but it raises a very important issue for EPs. Essentially, patients can have infection proximal to an obstructing stone and not have it suggested or proven by analyzing the urine from a Foley catheter or a clean-catch specimen. In the absence of a frank bacteremia, usually evident by a rather sick patient, such infections will not be suspected, and certainly many cannot even be proven.
So what is an emergency physician to do? First, it seems reasonable to invoke the caveat that all patients with a kidney stone should have a CT scan not only to confirm the diagnosis but to assess the degree of obstruction. A routine urinalysis also is suggested, with careful attention to signs of infection, not just the presence of blood. A CT scan likely will not demonstrate complete versus partial obstruction, but significant hydronephrosis/hydroureter certainly gets one's attention. Severe or complete obstruction is rarely clinically silent, and is obvious on the CT. And it rates a call to the urologist from the ED.
Many small stones produce minimal, inconsequential hydronephrosis. If the stone is larger than 5 mm, a size unlikely to pass spontaneously or readily, and there is hydronephrosis or hydroureter, it seems like same-day urologic consultation or admission is reasonable. Merely finding some degree of urinary obstruction is not an absolute indication for admission. It takes a few weeks of complete obstruction to produce kidney damage. I don't know whether hydronephrosis is an absolute indication for antibiotic prophylaxis — apparently it is not — but the issue is not defined in the literature and experience would suggest that for most patients with short-term partial obstruction, there is no mandate for immediate antibiotics. In my experience, infection is simply not an issue in most cases of renal colic.
The presence of a nonobstructing stone in the ureter does not necessitate antibiotics unless there are extenuating circumstances such as fever, pyuria, or subtle signs of infections (tachycardia, tachypnea, leukocytosis, the omnipresent chills, or a good clinical gestalt in patients at high risk). Patients with diabetes or other forms of immunocompromise should be approached quite carefully when one is considering discharging them with a kidney stone. It's prudent, although not universal, to have such high-risk patients admitted to the hospital. At least place a call to the urologist, and document the details of the discussion. In questionable patients, submitting urine for culture and sensitivity (for what it's worth) and having a low threshold for sending blood cultures is prudent.
You don't have to work too many shifts to understand that infection in a patient with an obstructing stone, particularly if he has diabetes or other immunocompromise, can quickly add up to a disaster. Such patients can become septic within a few hours, often right under your nose in the ED. While a run-of-the-mill small kidney stone is merely a major annoyance to most patients, one has to respect larger stones, those causing obstruction, and stones in individuals with the potential to rapidly decompensate from infection. Infection behind an obstructing stone is similar to an abscess. It has to be treated aggressively and often surgically drained with the requisite accuracy.
Emergency Drainage or Urosepsis Associated with Upper Urinary Tract Calculi Koji Y J Urol 2005;173(2):458
The authors of this study state a very obvious fact: Urinary tract infection associated with obstructive uropathy can result in bacteremia and sepsis. This study retrospectively reviewed and identified the clinical characteristics of patients undergoing emergency urinary drainage for urosepsis associated with urinary tract calculi. Patients eligible for review had sepsis, defined as fever, tachycardia, tachypnea, leukocytosis, positive urine culture, and significant pyuria, and were absent a clinical focus of infection outside the urinary tract.
Of the 424 patients reviewed, 84 had an emergency drainage procedure, generally a percutaneous nephrostomy or urethral stenting. The procedure was performed primarily for established urosepsis, but also for intolerable pain or renal failure due to bilateral obstruction. About a quarter of the patients were sick enough to require vasopressors, and they suffered multisystem involvement. The mortality rate was two percent. Overall, about 20 percent of patients with calculi required an emergency drainage procedure for urosepsis. That means that 80 percent of infected patients did not require immediate surgical intervention, but criteria for surgery were not defined by the authors. About 20 percent of the patients had risk factors such as diabetes, another immunocompromise, or old age.
The most common stone was calcium phosphate/oxalate, and the most commonly isolated urinary uropathogen was E. coli. The authors note another interesting theoretical infection-related risk factor to obstructive uropathy: increased intrarenal pelvic pressure secondary to an obstructing stone decreases the delivery of drugs, such as antibiotics, to the kidneys. That concept is not well promulgated.
Comment: This article suggests that patients with a urinary tract infection in the presence of calculi (unclear whether there was complete obstruction) will require emergency drainage for urosepsis about 20 percent of the time. Not unexpectedly, older patients and those with immunocompromise are at higher risk for developing urosepsis. This paper does not specifically help emergency physicians identify patients at high risk, but it does raise the specter of potential disasters from stones and urinary tract infections. The takeaway point from this article for me is that patients with signs, symptoms, or suggestions of UTI in the presence of a stone or patients with a stone who are at high risk for infection in general should be approached with great clinical caution. This will include an analysis for subtle clinical signs of infection, careful attention to follow-up, and a low threshold for admission and urological consultation. The emergency physician cannot be satisfied simply by making a brilliant diagnosis of a kidney stone and loading the patient with morphine. That's merely the first chapter in the book on emergency medicine and kidney stones.
Emergency physicians occasionally see patients after lithotripsy with pain from the passage of calculi or signs and symptoms of UTI. While the incidence of urinary tract infection is low following lithotripsy, especially with sterile urine beforehand, some broken and infected stones can precipitate UTI.
The optimal method for the urgent decompression of the collecting system for obstruction and infection due to urethral calculi is the subject of some debate among urologists. Common options include percutaneous nephrostomy or retrograde urethral catheterization. Neither modality has been demonstrated superior in promoting a more rapid recovery after drainage, although the concept of the need for one of the procedures is well supported. (J Urol 1998;160:1260.) It is generally considered standard to provide perioperative antibiotic prophylaxis during stone removal. (Eur Urol 2003;44:115.) Generally a fluoroquinolone is administered, and a small oral dose appears to be adequate for uncomplicated cases. Some urologists, however, still prefer the intravenous route.
The topic of an “infected stone” is somewhat of an enigma to most emergency physicians. The most common calcium oxalate stones are not caused by infection, nor do these stones harbor bacteria. Approximately 15 percent of kidney stones, however, harbor an infection inside the calculus and are called infectious stones. Expect a question about this on your next board exam: These infected stones are usually composed of struvite and/or carbonate apatite. These are unusual stones, and are not the standard calcium-containing stones that make up the majority of calculi that bring patients to the ED with renal colic.
Struvite and apatite stones form in urine that has become supersaturated by the byproducts of the hydrolysis of urea. The precondition for the formation of an infectious stone is the presence of a urease-positive urinary tract infection. Urease is an enzyme that splits urea into ammonia, and the resulting ammonia ions produce alkaline urine, a precondition for the formation of the struvite and carbonate apatite crystals. Crystals then coalesce to form stones, typically in the renal pelvis. They grow there, often rapidly, into the classic staghorn calculi. A tip-off to these stones is highly alkaline urine in the presence of infection.
Most emergency physicians don't look at the pH of the urine in patients with renal calculi, but would look like true academicians if they did. The organisms most frequently cultured from patients with these struvite stones are Klebsiella and Proteus mirabilis. E. coli and Pseudomonas also can be found. The problem with infected stones is that the calculi function as a sanctuary for organisms, protecting them from eradication by antibiotics.
Emergency physicians should be familiar with staghorn calculi. These can be seen quite easily on a plain abdominal x-ray, but usually are an incidental finding. They do not produce symptoms unless a small piece breaks off and passes into the ureter, becoming a standard kidney stone, or unless there is an overt infection. Over the years, they can lead to renal failure, particularly if they are present bilaterally. Staghorn calculi also may be a focus for recurring infection. If you find a staghorn calculus as an incidental finding on your investigation for some other process, the patient should be informed of its presence and urological consultation should be routine.
Urinary Tract Infection and Renal Calculi
Problem: Most patients with acute renal colic do not have concurrent urine infections. Staghorn calculi are the exception, and these stones are actually produced in persistently alkaline urine by chronic infection with urea-splitting organisms. It is not standard to order a urine culture or prescribe prophylactic antibiotics for uncomplicated patients who are passing a renal calculus. If the stone produces complete obstruction, there is a potential for infection in stagnant urine collecting behind the stone. Some degree of obstruction is common yet inconsequential as the patient passes a stone.
Clinical Approach: Patients with obstructing calculi and those with immunocompromise are at highest risk for urosepsis. In complete obstruction, clinical deterioration can be rapid. A CT scan can identify hydroureter/hydronephrosis, but cannot predict partial vs. complete obstruction. Take the patient's temperature, and order a standard dipstick urinalysis to check for pH and leukocyte esterase/nitrites, which are suggestive of infection. Consider sending a microscopic analysis to look for leukocytes, and check a CBC for leukocytosis. A urine culture may not be sensitive for UTI proximal to an obstructing stone.
Intervention: Consider antibiotics and close follow-up if the vital signs, urinalysis, CBC, and clinical scenario suggest subtle or subclinical infection. Patients with obvious clinical infection with a stone are candidates for antibiotics. Those with obstruction or immunocompromise deserve initial IV antibiotics, admission to the hospital, and expeditious urological consultation. Surgical intervention may be required in the presence of urosepsis and obstruction.
Struvite and Staghorn Calculi
- ▪ Branched stones may fill all or part of the renal pelvis, and may branch into several or all calices.
- ▪ Most are composed of struvite (magnesium ammonium phosphate) and/or calcium carbonate (apatite).
- ▪ These calculi are often referred to as “infection stones” because they are associated with urinary tract infections from urea-splitting organisms (Proteus/Klebsiella).
- ▪ Bacteria create persistently alkaline urine, promoting stone growth.
- ▪ These can grow rapidly over a period of weeks to months, and if untreated can lead to decreased kidney function and end-stage renal disease.
- ▪ The calculi predispose patients to sepsis and recurrent infections.
- ▪ Most patients require definitive surgical treatment, although not always emergent.
- ▪ Bacteria live within the stone, making it difficult for antimicrobial agents to penetrate.
- ▪ Chronic antibiotic therapy may limit stone growth.
- ▪ Classis symptoms of nephrolithiasis are uncommon, but patients may present with UTI, flank pain, or hematuria.
- ▪ Urine pH is consistently alkaline (greater than 7.0) often with magnesium ammonium phosphate crystals in the urine sediment.
- ▪ Over a period of years, there is often a progressive deterioration in renal function, and elective removal of these stones is suggested.
- ▪ These stones are radio-opaque, and easily seen on a plain abdominal radiograph.
Reader Feedback: Readers are invited to ask specific questions and offer personal experiences, comments, or observations on InFocus topics. Literature references are appreciated. Pertinent responses will be published in a future issue. Please send comments to firstname.lastname@example.org. Dr. Roberts requests feedback on this month's column, especially personal experiences with successes, failures, and technique.
Dr. Roberts: I have questions regarding your recent article, “Kidney Stone or Aortic Catastrophe?”(EMN 2006;28:31.) Do you scan patients who have a CT scan documenting a prior kidney stone (and documenting the absence of an abdominal aortic aneurysm)? If you don't, how long will you rely on a prior negative CT scan? Do you rescan if the CT is a year old? Two years old? Three years old? If the CT without contrast is negative, do you rescan with contrast to look for aortic dissection? If you do sometimes, how do you decide who needs a scan with contrast? What is the role of bedside ultrasound to exclude AAA, and perhaps save the patient radiation and money? — Andrew McVie, MD
Dr. Roberts responds: Dr. McVie's comments highlight the diversity of clinical approaches to suspected renal colic. There are no specific standards or guidelines, so my answers are from my perspective. I like to know as much as possible about such patients, and that includes size and position of the stone, degree of obstruction, and associated or unsuspected pathology. I usually do a noncontrast CT on all patients, even when I am sure they have a stone, and even if they had a CT a month ago for another stone. That approach is not universally accepted, and maybe I'm just paranoid or insecure.
If an aortic aneurysm was absent two to three years ago, there is no reason to repeat the CT with the sole intent of looking for a new aortic aneurysm. Aortic disease is tricky, and dissection/aortic occlusion can occur suddenly, while aneurysms usually don't grow that fast. In patients at high risk for dissection (hypertension, diabetes, cocaine use, other vascular disease, etc.), each bout of probable renal colic is a new ballgame. If you have not seen an aortic dissection in a 35-year-old, you just have not worked enough shifts.
The added value of IV dye for any CT is overrated so don't buy the radiologist's lament that it's a worthless study for aortic disease if you don't use contrast. A good radiologist should get a pretty good look at the aorta with a noncontrast study. If you are accomplished with the bedside ultrasound, go for it; we old guys will never be facile with that fancy machine when it comes to subtle aortic disease, but we also know for sure that the ED is no place to try to save money. Radiation is certainly of concern; I don't have an answer for that. Like most things in medicine, these issues don't matter much if you are right, and there are no subsequent problems. I'm not that smart, so I evoke that sentiment of Clint Eastwood in Dirty Harry, “Are you feeling lucky?”