| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, University of Alabama, 619 S. 19th St. Birmingham, AL
35233.
2
Department of Biostatistics, Kansas Cancer Institute, 3901 Rainbow Blvd.,
Kansas City, KS 66160-7312.
3
Department of Surgery, The Trauma Center of the University of Alabama,
Birmingham, AL 35233.
Received March 23, 1999;
accepted after revision June 15, 1999.
Address correspondence to D. E. Morgan.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Hemodynamically stable trauma patients with hematuria were examined under standardized protocol with dynamic oral and IV contrast-enhanced CT of the abdomen and pelvis, followed immediately by CT cystography. CT cystography consisted of contiguous 5-mm axial scans of the pelvis after retrograde distention of bladder with 300-400 ml of 4% iodinated contrast material. Radiographic and clinical variables (pelvic fracture, pelvic fluid, intraabdominal visceral injury, degree of hematuria, hematocrit, units of blood transfused, base deficit, injury mechanism, seat belt use, sex, age) were assessed and statistically analyzed using the two-tailed Fisher's exact test and Wilcoxon's rank sum test. Positive and negative individual and multivariate predictors were analyzed.
RESULTS. Of the 157 patients entered in our study, 12 (eight males and four females) had bladder rupture. One or more pelvic fractures were present in nine (75%) of the 12 patients (p < 0.001). Pubic symphysis diastasis, sacroiliac diastasis, and sacral, iliac, and pubic rami fractures were statistically associated with bladder rupture. Isolated acetabular fractures did not correlate with rupture. Eight (67%) of the 12 patients with bladder rupture revealed on CT cystography had gross hematuria (p < 0.001). No ruptures were seen in patients with <25 RBC/HPF (red blood cells per high-power field). All patients with rupture had pelvic fluid revealed on standard contrast-enhanced CT (p < 0.001).
CONCLUSION. Gross hematuria, pelvic fluid, and specific pelvic fractures were highly correlated with bladder rupture; identification of these findings may help in selection of trauma patients for CT cystography.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In the past, the standard examination of trauma patients with suspected urinary tract injury included IV pyelography and conventional cystography [2] and, in men, retrograde urethrography. Today, hemodynamically stable trauma patients are routinely examined with contrast-enhanced abdominopelvic helical CT. This technique readily reveals renal injuries as well as injury to other abdominal viscera. Contrast-enhanced CT with delayed imaging [5] or retrograde bladder filling [6, 7, 8] has been shown to be highly accurate and equivalent to conventional cystography in the detection of bladder rupture. This method, commonly referred to as CT cystography, is often easier to perform than conventional cystography in patients who have sustained severe trauma with complex pelvic fractures.
Over an 18-month period, the trauma service at our institution followed a policy that all trauma patients with any degree of hematuria would undergo full evaluation of the urinary tract consisting of contrast-enhanced CT followed by CT cystography. This policy produced a data set that allowed us to study consecutive patients regardless of the degree of clinical suspicion for injury. The objective of this study was to prospectively evaluate clinical and radiographic variables that correlated with positive findings on CT cystography. These variables could then be used for selection criteria in determining which trauma patients should in the future be evaluated with CT cystography after standard contrast-enhanced CT.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A work sheet was filled out at the time of the CT by the on-call senior radiology resident or attending abdominal imager. All scans were reviewed the next morning during a daily CT quality assurance conference. Radiographic variables evaluated included pelvic fractures, specifically pubic rami, iliac, sacral, and acetabular fractures, and pubic symphysis and sacroiliac diastasis. Other urinary tract injuries including urethral injuries (demonstrated on retrograde urethrography) and renal injuries (demonstrated on CT) were documented. Traumatic injury to other abdominal viscera and the presence of pelvic fluid (defined as intra- or extraperitoneal hematoma, or free pelvic fluid) on standard contrast-enhanced CT were noted. Postdrainage findings in the pelvis were recorded.
We evaluated the following clinical variables: sex, age, mechanism of injury, seat belt use, base deficit, degree of hematuria, units of blood transfused, and hematocrit. The degree of hematuria was determined by visual inspection and laboratory analysis and was classified into one of the following categories: gross hematuria or microscopic hematuria (0-11, 11-25, or >25 [too numerous to count] RBC/HPF [red blood cells per high-power field]). The mechanism of injury was categorized as either a diffuse mechanism, such as motor vehicle collision, or as a focal mechanism, including fall, assault to pelvic region, and penetrating injury to pelvic region. The base deficit, which reflects the metabolic component of a patient's acidosis, was calculated according to a standard nomogram using serum bicarbonate, partial pressure of carbon dioxide, and pH obtained on routine arterial blood gas. Statistical analysis of each variable and several combined variables was performed using the two-tailed Fisher's exact test and the Wilcoxon's rank sum test.
Although it was not an objective to evaluate the accuracy of CT cystography as part of our study, interpretation of the CT cystography was correlated with operative findings, findings from subsequent radiographic evaluations, or clinical follow-up (trauma registry and trauma clinic records review) in all patients.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Forty-four (28%) of our 157 patients had single or multiple pelvic fractures (Table 1). Nine (75%) of 12 patients with bladder rupture had one or more pelvic fractures (p < 0.001). Eight (89%) of nine patients with extraperitoneal rupture had one or more pelvic fractures (Fig. 1A, Fig. 1B). Two patients with combined intra- and extraperitoneal rupture had pelvic fractures. Two patients with isolated intraperitoneal rupture and one patient with extraperitoneal rupture had no fractures. Specific types of pelvic fractures correlated statistically with bladder rupture. Pubic symphysis diastasis (p < 0.001), sacral fracture (p < 0.001), pubic rami fractures (p < 0.001), sacroiliac joint diastasis (p < 0.001), and iliac fracture (p < 0.031) correlated positively with bladder rupture. Isolated acetabular fractures did not correlate statistically with bladder rupture (Fig. 2A, Fig. 2B).
|
|
|
|
|
On standard abdominopelvic contrast-enhanced CT, all patients with bladder rupture demonstrated free fluid and/or hematoma in the pelvis (p < 0.001) (Fig. 3A, Fig. 3B). Twenty-eight (19%) of 145 patients without bladder rupture had free fluid. Other urinary tract injuries detected in the patients with bladder rupture included one patient with a renal laceration (detected by contrast-enhanced CT) and one with a posterior urethral injury (detected by retrograde urethrography). There was no higher incidence of injuries of the solid viscera elsewhere in the abdomen in patients with bladder rupture compared with patients without bladder rupture. No additional cases of bladder rupture were found on the postvoid series.
|
|
We analyzed combinations of clinical and radiographic variables. Positive multivariate predictors (Table 2) for bladder rupture included the combination of pelvic fluid and gross hematuria, where bladder rupture was found in eight (67%) of the 12 patients with these predictors; pelvic fracture and gross hematuria, where bladder rupture was found in six (60%) of the 10 patients with these findings; and combined pelvic fracture, pelvic fluid, and gross hematuria, where bladder rupture was present in six (67%) of nine patients with these three findings. Negative multivariate predictors (Table 3) included the following: the combination of microhematuria (<25 RBC/HPF) and absence of pelvic fluid on standard abdominopelvic contrast-enhanced CT, where no bladder rupture was present in any of the 33 patients with these findings; the combination of absence of pelvic fracture and absence of gross hematuria (no bladder rupture in 99 of 99 patients); and the combination of absence of pelvic fracture and absence of fluid on standard abdominopelvic contrast-enhanced CT (no bladder rupture in 100 of 100 patients) (Fig. 4A, Fig. 4B). These multivariate trends did not reach statistical significance.
|
|
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Previous experience has shown the combination of IV pyelography and conventional cystography, utilized in the past, was of low yield if all patients with any degree of hematuria underwent diagnostic testing [2]. We similarly hypothesized that not all trauma patients with any degree of hematuria should require CT cystography after standard abdominal contrast-enhanced CT. The purpose of this study was to prospectively evaluate clinical and radiographic variables that correlate with positive findings on CT cystography and then use these criteria to select the subset of trauma patients in whom CT cystography would be most beneficial. We did not attempt to restudy the accuracy of CT cystography compared with standard cystography or surgical findings as part of this study. Based on operative findings and clinical follow up, however, no bladder ruptures or urethral injuries were missed in our patient population, nor were any cases proven to be false-positives.
In this study, gross hematuria was highly correlated with bladder rupture. All 12 patients with bladder rupture had either gross hematuria (n = 8) or >25 RBC/HPF (n = 4). Similarly, in the series of 25 patients with proven urinary tract trauma [3], all 10 patients with injury to the lower urinary tract had gross hematuria. In the large series of patients with bladder injuries described by Cass [1], most (132/165 [80%]) of the patients with rupture had gross hematuria; however, six patients with <30 RBC/HPF had bladder rupture, and 16 patients without hematuria had bladder rupture. This discrepancy with our results—of no bladder rupture being present in patients with <25 RBC/HPF—is difficult to resolve. The difference may, in part, be a result of the trauma population selected or the method of obtaining specimens. The possibility of bladder rupture in patients with lesser degrees of hematuria prompted an evaluation of microscopic hematuria combined with other variables in our population. No bladder rupture was found in 123 patients with microscopic hematuria or the absence of free fluid on standard contrast-enhanced CT. No bladder rupture was detected in 99 patients with the combination of microscopic hematuria and absence of pelvic fracture. Furthermore, no bladder ruptures were detected in the 10 patients with one or more pelvic fractures with only microscopic hematuria (<25 RBC/HPF).
Pelvic fracture is a common associated injury in patients with bladder rupture, reported in 75-83% of cases [1,9]. In our series, pelvic fracture was present in nine (75%) of 12 patients with bladder rupture. Flaherty et al. [10] reported that public arch fractures, especially bilateral, were more likely to cause or be associated with bladder rupture. Cass [1] reported that 55% of all patients with bladder rupture and pelvic fracture had pubic rami fractures, making this fracture the most frequent type of pelvic fracture in that series. In our study, pubic rami fractures and pubic symphysis diastasis were highly correlated with bladder injury (p < 0.001 for both). Sacral fracture, sacroiliac diastasis, and iliac fracture also correlated positively with rupture (p < 0.001, p < 0.001, p < 0.031, respectively). Isolated acetabular fracture was the only pelvic fracture that did not have a statistically significant correlation with bladder rupture in our population. Conversely, evaluation for bladder rupture in all patients with pelvic fractures is not prudent because the incidence of injury to the lower urinary tract in patients with pelvic fracture as the index lesion is only 8% (range, 3.3-16%) [1]. It is also important to remember that the absence of pelvic fracture does not exclude bladder rupture, especially intraperitoneal rupture. According to the literature, pelvic fracture may be absent in 17-25% of patients with bladder rupture [1, 9]. Three (25%) of 12 patients with rupture in our series had no pelvic fracture on conventional radiographs or review of bone windows on contrast-enhanced CT scans; of these patients, two had intraperitoneal rupture. In multivariate analysis, the absence of pelvic fracture may be useful. In our study, if there was no pelvic fracture and no free fluid on standard contrast-enhanced CT, the bladder was intact in all of the 100 patients with this combination of features. Likewise, the bladder was intact in all 95 patients who had no fracture combined with the absence of gross hematuria.
Hypotension was a clinical variable used in the past to determine which blunt trauma patients might have significant urinary tract injury. The absence of shock in patients with blunt or penetrating trauma and microhematuria is a very strong negative predictor for renal trauma [2, 3]. In our study, there was no statistically significant difference in blood pressure between those patients with bladder rupture and those without. This could be due in part to better resuscitation techniques than in years past or more likely reflects the hemodynamically stable trauma population examined using abdominal CT.
The difference in the base deficit of patients with bladder rupture compared with that of patients without rupture was significant. Base deficit is a general indicator of injury severity and becomes more negative when tissue hypoperfusion has occurred. Regarding the mechanism of injury, focal pelvic trauma was statistically associated with bladder rupture (p < 0.04), but blunt trauma was not, even though most patients (8/12) with bladder rupture were injured in a motor vehicle accident. This is a result of the large number of patients with blunt trauma who did not have a bladder rupture. Note also that all patients with focal pelvic trauma and bladder rupture had gross hematuria. Other clinical variables measured in this study (age, sex, hematocrit level, number of units of blood transfused, and seat belt use) had no statistical correlation with bladder rupture.
In addition to the specific pelvic fractures that correlated highly with bladder rupture, other radiographic variables evaluated in this study were important in predicting bladder injury. All patients with bladder rupture, whether extraperitoneal or intraperitoneal or both, had pelvic fluid on standard abdominal contrast-enhanced CT. Of the 145 patients in our study who did not have bladder rupture, 28 (19%) had pelvic fluid. There was no higher rate of concomitant injury to intraabdominal viscera in our patients with bladder rupture to explain the increased occurrence of free fluid. The absence of fluid was a strong negative predictor. In their initial experience with standard contrast-enhanced CT detection of bladder rupture, Mee et al. [11] also suggested that the absence of pelvic fluid (either intra- or extraperitoneal) indicated the absence of rupture. This correlation with pelvic fluid makes intuitive sense. In patients with bladder rupture, urine or hematoma or both could be appreciated in the region of the bladder. Complete absence of any abnormal fluid would be unlikely in the presence of a bladder rupture. In the absence of pelvic fluid, especially if only microscopic hematuria is present, our data suggest that CT cystography may be avoided. The presence of fluid, however, is not as strong a positive predictor because hematoma can result from orthopedic injury alone. Based on these data, CT cystography should be done when pelvic fluid is present, especially when there are fractures or gross hematuria, to define which of the patients has a bladder rupture and to define the type (intraperitoneal or extraperitoneal).
Other urinary tract injuries detected in our patients with bladder rupture included one renal injury and one posterior urethral injury. This is less than the series of Cass [1], in which posterior urethral injury was noted in 17% of patients with bladder rupture. Patients suspected of having posterior urethral injuries (those with blood at the urethral meatus or gross hematuria) underwent retrograde urethrography before insertion of a Foley catheter or pericatheter retrograde urethrography after CT cystography. This examination was clinically effective in excluding urethral injuries in the patients with bladder rupture in our study. The negative findings on retrograde urethrography in seven patients, as well as the lack of injury in the remaining four, were confirmed with clinical follow-up.
Historically, a lack of a radiograph obtained after bladder drainage [12] or a lack of a cystogram obtained when the bladder was fully distended [1,2,5,6 (or both) have led to false-negative interpretations of standard cystograms. The lack of full bladder distention using antegrade filling techniques during CT cystography also resulted in early false-negative reports [11], although a later study using this same technique showed accuracy equivalent to that of standard cystography [5]. Performing CT cystography with retrograde bladder filling to maximum capacity alleviates the problem of inadequate distention [7,8,9 and generally requires 300-400 ml of contrast medium. With cross-sectional imaging of CT cystography, the distended bladder does not obscure extravasation of contrast medium from the bladder. We evaluated the additional information obtained from the postvoiding imaging series through the pelvis after CT cystography and found no cases in which bladder rupture was visualized on only the postvoid series. We have eliminated this series from our protocol.
This study has several potential limitations. The individual variables measured in this study achieved statistical significance using the two-tailed Fisher's exact test; however, the number of positive cases did not provide statistical significance with multivariate analysis, which might have been possible with a larger series. Surgical proof of bladder rupture was obtained in only seven of 12 patients with positive findings on CT cystography. There was not uniform surgical proof of positive cases because most of our patients had extraperitoneal ruptures, and these may be treated nonoperatively at our institution. No additional or missed ruptures were discovered, however, either by clinical or radiographic follow-up. Finally, the patients in our study were, by definition, hemodynamically stable, and our results may not be transferable to all trauma patients.
Several individual clinical variables, including gross hematuria and base deficit, and individual radiographic variables, including specific types of pelvic fracture and pelvic fluid, on standard contrast-enhanced CT had high statistical correlation with bladder rupture shown on CT cystography. Multivariate trends may also suggest which patients are likely to have bladder rupture. In our study, the patients most likely to have bladder rupture on CT cystography were those with pelvic fluid and gross hematuria (eight [67%] of 12 patients); those with pelvic fracture and gross hematuria (six [60%] of 10 patients); and those with a combination of pelvic fracture, pelvic fluid, and gross hematuria (six [67%] of nine patients). Likewise, multivariate findings may strongly suggest the absence of bladder rupture and can judiciously be used to exclude some trauma patients from unnecessary CT cystography and its associated costs.
We recommend CT cystography after abdominopelvic contrast-enhanced CT in the following hemodynamically stable trauma patients: patients with gross hematuria, patients with a pelvic fracture (other than an acetabular fracture) and high-level (>25 RBC/HPC) microhematuria, and patients with microhematuria and pelvic fluid. We suggest that CT cystography may be avoided in the following patients: patients with microhematuria (<25 RBC/HPC) and no pelvic fracture, patients with microhematuria (<25 RBC/HPC) and isolated acetabular fracture, and patients with microhematuria (<25 RBC/HPC) and no pelvic fluid. A postvoid series provides no additional information to the CT cystogram and should be omitted.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  C. J. Sivit Imaging Children with Abdominal Trauma Am. J. Roentgenol., May 1, 2009; 192(5): 1179 - 1189. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T M Wah and J A Spencer CT of adult urinary tract trauma Imaging, August 1, 2005; 17(1): 53 - 68. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
