| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
2000 ARRS President's Award |
1
Department of Radiology, University Hospitals of Cleveland, Case Western
Reserve University, 11100 Euclid Ave., Cleveland, OH 44106.
2
Department of Psychology, Case Western Reserve University, Cleveland, OH
44106.
Received February 24, 2000;
accepted after revision May 31, 2000.
Presented at the annual meeting of the American Roentgen Ray Society,
Washington, DC, May 2000.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Questionnaires were completed on 93 patients who were referred to the radiology department with clinically suspected renal colic. We prospectively surveyed the clinician's diagnostic confidence and treatment plan before and after unenhanced abdominal and pelvic CT.
RESULTS. Fifty-six patients (60%) had positive findings for calculi, 20 patients (22%) had normal findings, and alternative diagnoses were found in 17 patients (18%). The clinician's diagnostic certainty of stones before CT was variable with the largest frequencies at 41-60% (n = 30) and 71-90% (n = 35). The diagnostic certainty of stones after CT showed movement toward either less than or equal to 10% (n = 25) or greater than or equal to 91% (n = 51). The mean change in diagnostic confidence was 34%. Fifty-seven patients (61%) had a change in treatment plan. Specifically, the need for urology consultation as the initial treatment plan was reduced from 24 patients to one patient. Plans for admissions suggested before CT (n = 11) were nearly cut in half (n = 6) after imaging. Lastly, seven patients who would have initially been discharged were admitted to the hospital after imaging.
CONCLUSION. CT significantly increased emergency department clinician diagnostic confidence and altered initial treatment decisions in patients with suspected renal colic. Most often, CT confirmed a ureteral stone and allowed appropriate discharge or urologic intervention. In a smaller subset of patients, CT established a significant alternative diagnosis that allowed the prompt initiation of appropriate treatment.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Several studies have shown unenhanced helical CT to be the study of choice in the evaluation of patients with suspected renal colic [1,2,3]. Although CT has a high sensitivity and specificity in the detection of ureteral calculi, important alternative diagnoses can also be made [4]. Added benefits of this technique include elimination of IV contrast administration and a rapid scan time. Despite these results, there are little data addressing the impact of this modality on diagnostic thinking of clinicians and on patient therapy.
Fryback and Thornbury [5] and Thornbury et al. [6] described a hierarchic model to assess the contribution of a diagnostic test to patient care. We used their model to evaluate two intermediate levels of efficacy for unenhanced helical CT in the setting of acute flank pain in the emergency department—diagnostic thinking and therapeutic efficacy.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Subjects
We undertook a prospective study of 93 emergency department patients with
clinically suspected renal colic who underwent abdominal and pelvic CT at our
institution from February 1999 through July 1999. Each patient was examined by
an emergency department clinician who determined that abdominal and pelvic CT
was indicated for further evaluation on the basis of the patient's clinical
and laboratory presentation. Inclusion criteria were patients of all ages whom
emergency department clinicians determined had possible renal colic. This
study was approved by the institutional review board. Informed consent from
the participating clinicians was implicitly obtained when they agreed to
complete each survey.
Data Collection
Before initiating the study, the emergency department clinicians and
radiology residents were informed of the questionnaire procedure. The
importance of including all patients was stressed. The preexisting CT logbook
kept by CT technologists was reviewed on a weekly basis to assess compliance
with data collection. This logbook lists all patients, their study type,
indication, date, and requesting department.
The emergency department clinician who requested the CT completed a verbal questionnaire on all patients before and after CT results were revealed. The pre-CT questionnaire was completed after the emergency department clinician performed clinical and laboratory examinations but before the availability of CT results. The objectives of the pre-CT questionnaire were to determine the clinician's diagnostic confidence (expressed as a percentage, 0-100%) that renal and ureteral calculi were the cause of the patient's symptoms and to obtain data on the anticipated initial treatment plan. Survey treatment plans were hospital admission, discharge, observation, urology consultation, and other consultations.
After CT, the same emergency department clinician was contacted and informed of the imaging results. The clinician was then asked to provide a revised level of diagnostic confidence that renal calculi were the cause of the patient's presentation (0-100%) and a revised anticipated treatment plan with the same measures used before the pre-CT questionnaire. In addition, the clinician was asked to indicate the primary use of the recent CT. Most surveys were completed by emergency department attending radiologists, but also by residents. The questionnaires were administered by the radiologists who interpreted the CT, including residents and attending radiologists during daytime hours and residents at night. Unenhanced CT is the primary imaging modality used by our emergency department in the evaluation of patients with suspected renal colic.
CT scans were prospectively evaluated for the presence of renal, ureteral, and bladder calculi. Secondary signs including hydronephrosis, hydroureter, perinephric stranding, "tissue-rim sign," and flattening of the adjacent bladder wall were noted [7]. Positive findings were defined by the presence of renal, ureteral, or bladder calculi or by the presence of one or more secondary signs of obstruction with a calcification in the expected course of the ureter. Normal findings on examination indicated absence of renal, ureteral, or bladder calculi and absence of secondary signs of obstruction. Alternative diagnoses on CT were noted.
Patient outcomes were determined by review of emergency department records, operative reports, hospital discharge summaries, hospital charts, and the hospital's computerized clinical database. Patient data collection included patient age and gender, chief complaint, history of nephrolithiasis, urinalysis, radiology studies, urology consultation, and interventional procedures.
Most patients underwent helical CT (PQ5000 scanner; Picker International, Cleveland, OH) with 5-mm collimation at 5-mm intervals from the superior aspect of the kidneys to the pubic symphysis in the supine position. All patients were initially imaged without contrast administration. After the unenhanced studies, four patients required IV contrast administration as a problem-solving technique; each of them had an alternative diagnosis before contrast administration. CT scans were reviewed with a window level and width appropriate for soft tissue (window, 325 H; level, 35 H) on hard copy.
Data Analysis
Efficacy of diagnostic thinking.—The impact of CT
information on the referring clinician's diagnosis was assessed by evaluating
the percentage point change in diagnostic certainty between pre- and post-CT
questionnaires. We used the methods proposed by Fryback et al.
[5] and Thornbury et al.
[6] and applied them to similar
intermediate imaging outcome studies
[8,9,10].
When the diagnosis was the same before and after CT, the change in diagnostic confidence was calculated by subtracting the percentage of diagnostic certainty before CT from the percentage of certainty after CT. For example, if the diagnostic certainty of renal stone disease before CT was 75% and CT supported this diagnosis and hence raised diagnostic certainty to 100%, then the gain in diagnostic certainty from CT was 25% (100% - 75% = 25%).
If the diagnosis changed before and after CT, it was necessary to infer a diagnostic confidence level before CT for the final diagnosis. The sum probabilities of all diagnoses, including unstated diagnoses, was 100% [8]. If the pre-CT diagnostic confidence of renal stone disease was greater than or equal to 50%, then the inferred diagnostic confidence for the adjusted diagnosis was estimated as 100% minus the initial stated probability for renal stone disease. For example, if the emergency department clinician suspected renal colic with a pre-CT diagnostic confidence of 75% and appendicitis was identified on CT with a diagnostic confidence of 95%, the upper bound on pre-CT diagnostic confidence of appendicitis was 25% (100% - 75% = 25%). The gain in diagnostic confidence in this example would be 70% (95% - 25% = 70%).
If the diagnoses before and after CT were different and the diagnostic confidence for the incorrectly stated diagnosis before CT were less than 50%, then the inferred diagnostic confidence percentage was estimated as the same level given by the referring clinician. For example, if the emergency department clinician suspected renal colic before CT with a diagnostic confidence of 40% and CT revealed normal findings with a post-CT confidence of 95%, the gain in diagnostic confidence was estimated at 55% (95% - 40% = 55%). The 95% confidence intervals (CI) and analysis of variance were calculated for mean gain in diagnostic confidence (SPSS for Windows; SPSS, Chicago, IL).
Therapeutic efficacy.—The impact of CT on the initial treatment plan of emergency department clinicians was analyzed by comparing the pre- and post-CT treatment plans questioned on the survey. The percentage of patients in whom the treatment plan changed was calculated. In addition, the primary use of CT was assessed.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Table 1 describes the presenting clinical features, including hematuria, pain location, and history of renal calculi. Forty-eight patients presented with hematuria, of whom 35 (73%) had calculi on CT. Nineteen patients (40%) with calculi did not have hematuria at presentation. Thirty-three (94%) of 35 patients with a previous history of stone disease had positive CT findings for calculi. All patients (n = 24) presenting with hematuria and a history of renal calculi had positive CT findings for renal stones. A change in treatment plan was identified in 17 (71%) of these patients and included eight patients who were discharged from the emergency department instead of undergoing the pre-CT plan of urology consultation. No negative studies or alternative diagnoses were identified in this subset of patients.
|
Table 2 shows the diagnosis distribution and clinical outcome. Fifty-six (60%) of the original 93 patients had positive findings for renal calculi, 20 patients (22%) had normal findings, and alternative diagnoses were found in 17 patients (18%). Alternative diagnoses included appendicitis (n = 2), diverticulitis (n = 2), pancreatitis (n = 2), retroperitoneal hemorrhage (n = 2), pelvic inflammatory disease (n = 1), small-bowel obstruction (n = 1), pyelonephritis (n = 1), inflammatory bowel disease (n = 1), and a renal abscess (n = 1).
|
The overall gain in clinician diagnostic certainty was 34% (95% CI, 29-38%). Figure 1 shows the distribution of the pre- and post-CT percentage points representing clinician confidence in the diagnosis of renal stones. The clinician's pre-CT diagnostic certainty of stones was highly variable with the largest frequencies at 41-60% (n = 30) and 71-90% (n = 35). Patients with a prior history of calculi had a mean pre-CT diagnostic certainty of stones at 74% compared with 63% for the remaining sample. The post-CT diagnostic certainty of stones showed movement toward either less than or equal to 10% (n = 25, 27%) or greater than or equal to 91% (n = 51, 55%), indicating increased certainty that the patient either did or did not have a renal stone. A score of 0% indicated 100% confidence in an alternative diagnosis or normal study. The largest change in clinician diagnostic confidence resulted from those patients who had an alternative diagnosis on CT (mean change in diagnostic confidence, 55%; 95% CI, 46-64%; analysis of variance, p < 0.0001). In addition, a substantial change in diagnostic confidence was identified in the subset of patients with normal findings on CT (mean change in diagnostic confidence, 34%; 95% CI, 21-46%; analysis of variance, p < 0.01). The smallest change in clinician diagnostic confidence was identified in the group of patients with a positive finding on CT for calculi (mean change in diagnostic confidence, 26%; 95% CI, 21-31%). Table 3 shows the contribution of each group of patients (stone diagnosis, alternative diagnosis, and normal finding) to the overall change in diagnostic confidence.
|
|
A change in initial treatment plan was seen in 56 patients (61%) as a result of CT. Table 4 outlines the treatment changes. Specifically, seven patients would have initially been discharged before CT examination and, instead, were admitted to the hospital after imaging for the following reasons: four patients required procedures (3 urologic, 1 surgical; Figs. 2A,2B,2C and 3), two had significant retroperitoneal hemorrhages (Fig. 4), and one had CT findings suggestive of pancreatitis, subsequently confirmed with a finding of elevated serum amylase level.
|
|
|
|
|
|
After imaging, urology consultation as the initial treatment plan was reduced from 24 patients to one patient. For example, CT of a 55-year-old man with abdominal pain revealed findings of sigmoid diverticulitis. After CT, the patient was admitted and placed on IV antibiotic therapy instead of the pre-CT treatment plan of urology consultation. For 15 patients with the initial treatment plan of urology consultation, subsequent CT confirmed a small urinary calculus (Figs. 5A,5B,6A,6B,7). All were discharged from the emergency department with pain medication, instructed to strain their urine, and to follow up on an outpatient basis without immediate urology consultation.
|
|
|
|
|
Lastly, the number of admissions before CT (n = 11) was nearly cut in half (n = 6) after CT. Note should be made that one inpatient was included in the study but was excluded from this particular data set. However, CT in this patient was consistent with focal pyelonephritis, and antibiotic therapy was instituted.
Overall, emergency department clinicians stated that the primary use of CT was documentation of stone disease (n = 40). In patients with a positive finding for renal calculi, determination of stone size (n = 24) was indicated as the most important feature, outweighing stone location and evaluation for hydronephrosis.
Three-month clinical follow-up revealed no significant discrepancies on review of the hospital computerized clinical database and radiology database.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A six-tier hierarchal model of efficacy has been described by Fryback and Thornbury [5]. Using this model, one can ascertain the impact of an imaging test on patient outcome. The six levels of efficacy assessment are as follows: technical characteristics of an image, diagnostic accuracy, diagnostic thinking, therapeutic thinking, patient outcome, and societal effect. Inherent in this model is that an imaging study be efficacious at lower levels to be efficacious at higher levels.
Considerable research has documented the technical (level 1) and diagnostic accuracy (level 2) of unenhanced CT for the examination of patients with suspected renal colic [4]. Several studies have shown unenhanced helical CT to have a high sensitivity, specificity, and accuracy in the detection of ureteral calculi (level 2). For example, Smith et al. [4] found CT to have a sensitivity of 97%, a specificity of 96%, and an accuracy of 97%. However, little evaluation of its impact on clinician diagnostic confidence (level 3) and patient treatment (level 4) has been performed.
Improved diagnostic thinking efficacy refers to the situation in which an imaging study contributed to a diagnostic end point [5]. CT had the largest impact on clinician diagnostic confidence in patients initially suspected of having renal colic but who were found to have an alternative diagnosis. As expected, the smallest change in clinician diagnostic confidence was recognized in the group of patients with calculi identified on CT. By definition, referring clinicians ordered this specific imaging test because the pretest probability of renal colic was high.
Therapeutic efficacy refers to the impact of an imaging test on patient treatment [5]. CT altered patient treatment in 16 of 17 patients with an alternative diagnosis. In addition, a significant reduction in immediate urologic consultation was seen in 15 patients who were, instead, discharged after CT confirmed a small ureteral calculus.
CT has replaced the IV pyelogram as the imaging modality of choice in the evaluation of patients with renal colic. Nonetheless, it is not without limitations. Functional information, such as the amount of obstruction caused by a calculus, may be difficult to assess on unenhanced CT. However, obstruction is not the most important factor influencing the need for urologic intervention. Instead, stone size and patient symptomatology are more significant determinants dictating urologic intervention [11]. Our data confirmed that referring physicians cited determination of stone size as the most important use in positive findings. This clinical decision is supported by a recent report by Boulay et al. [12], who described conservative treatment in patients with a calculus measuring approximately 3 mm, whereas procedures such as stent placement or stone basketing were performed in patients with a calculus of approximately 7 mm or greater.
This study was initiated within 2 years of the introduction of this imaging modality to referring clinicians at our institution. Over time, one would expect an increase in use of this modality that may theoretically result in a smaller percentage of positive findings. A recent report by Chen et al. [13] described a trend in reduction of stone detection as clinicians became familiar with the CT technique. In our study, all patients (n = 24) presenting with hematuria and a history of renal stone disease had a positive finding for calculi. None had an alternative diagnosis. The value of CT in these patients is not in increased diagnostic confidence but in an accurate treatment plan. Review of the use of CT may serve its purpose by allowing clinicians to reexamine the importance of imaging this subset of patients immediately on presentation.
The limitations of this study include the general limitation of before—after studies in that what the clinicians report on the survey may not be accurate. Instead, clinical decisions may change during the course of patient care. In addition, our study is limited by our inability to measure out-of-system emergency department use and outpatient follow-up care. We could track follow-up visits to our emergency department, hospital admissions, and imaging at 3 months after the initial emergency department visit. A comparison of overall cost in patient care with CT versus other radiologic and nonradiologic tests remains to be determined.
In conclusion, CT significantly increased emergency department clinician diagnostic confidence and initial treatment decisions in patients with suspected renal colic. Most often, CT confirmed an ureteral stone and allowed appropriate discharge or urologic intervention. In a smaller subset of patients, CT established a significant and unexpected alternative diagnosis that allowed the prompt institution of an appropriate therapeutic regimen.
Acknowledgments
We thank Neil Abramson for his advice and editorial input.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Niemann, T. Kollmann, and G. Bongartz Diagnostic Performance of Low-Dose CT for the Detection of Urolithiasis: A Meta-Analysis Am. J. Roentgenol., August 1, 2008; 191(2): 396 - 401. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P.-A. Poletti, A. Platon, O. T. Rutschmann, F. R. Schmidlin, C. E. Iselin, and C. D. Becker Low-Dose Versus Standard-Dose CT Protocol in Patients with Clinically Suspected Renal Colic Am. J. Roentgenol., April 1, 2007; 188(4): 927 - 933. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Ng and C. Palmer Analysis of diagnostic confidence and diagnostic accuracy: a unified framework Br. J. Radiol., March 1, 2007; 80(951): 152 - 160. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Kirpalani, K. Khalili, S. Lee, and M. A. Haider Renal Colic: Comparison of Use and Outcomes of Unenhanced Helical CT for Emergency Investigation in 1998 and 2002 Radiology, August 1, 2005; 236(2): 554 - 558. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. S. Katz, M. Jain, M. J. Lane, E. M. Meiner, S. Bhalla, C. O. Menias, and C. M. Rucker Invited Commentary * Authors' Response RadioGraphics, October 1, 2004; 24(suppl_1): S28 - S33. [Full Text] [PDF]
|
|
|
|
|
|
D. Tack, V. De Maertelaer, and P. A. Gevenois Dose Reduction in Multidetector CT Using Attenuation-Based Online Tube Current Modulation Am. J. Roentgenol., August 1, 2003; 181(2): 331 - 334. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. P. Tamm, P. M. Silverman, and W. P. Shuman Evaluation of the Patient with Flank Pain and Possible Ureteral Calculus Radiology, August 1, 2003; 228(2): 319 - 329. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Tack, S. Sourtzis, I. Delpierre, V. de Maertelaer, and P. A. Gevenois Low-Dose Unenhanced Multidetector CT of Patients with Suspected Renal Colic Am. J. Roentgenol., February 1, 2003; 180(2): 305 - 311. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. H. Gottlieb, T. C. La, E. N. Erturk, J. L. Sotack, S. L. Voci, R. G. Holloway, L. Syed, I. Mikityansky, A. T. Tirkes, R. Elmarzouky, et al. CT in Detecting Urinary Tract Calculi: Influence on Patient Imaging and Clinical Outcomes Radiology, November 1, 2002; 225(2): 441 - 449. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Catalano, A. Nunziata, F. Altei, and A. Siani Suspected Ureteral Colic: Primary Helical CT Versus Selective Helical CT After Unenhanced Radiography and Sonography Am. J. Roentgenol., February 1, 2002; 178(2): 379 - 387. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
