Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 06/01/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
569
The Significance of Hematuria Children After Blunt Abdominal Trauma
H. Philip
Robert
Stalker1’2
A. Kaufman1’3’4 Kurt Stedje5
The clinical significance of hematuria in children who sustain blunt continues to be debated, as do the criteria for diagnostic imaging Previous reports have discussed the usefulness of certain clinical injury, such as the amount of hematuna present, the presence of injury,
and
the
presence
or absence
of symptoms
or findings
in
abdominal trauma in this population. predictors of renal shock or of head
on physical
examination.
To assess the value of such predictors of renal injury in children with posttraumatic hematuria, we reviewed and analyzed the medical records and abdominal CT examinations of 256 children with blunt abdominal trauma. One hundred six children (41%) had hematuria. Thirty-five patients (14%) had renal injury that could be diagnosed by using CT. Nine of these had clinically significant injuries according to our criteria. We found a direct relationship between the amount of hematuria and the severity of renal injury. Hypotension at presentation occurred in 38 patients and was an insensitive predictor of renal injury. The combination of hypotension and hematuna was no more sensitive than hematuria alone in predicting renal injury. Sixty patients had concomitant craniofacial injuries. This subgroup had the same prevalence of hematuria and renal injury as the group that did not have head injuries. There were no clinically occult renal injuries in the study population. Furthermore, we found that no normotensive child with fewer than 50 RBCs per high-power field had a significant renal injury, and conversely, all children with significant renal injuries had either large amounts of hematuna or shock.
154:569-571,
AJR
Hematuria Received August 14, 1989; accepted after revision October
24, 1989.
Presented
at the 31 st annual meeting
Society for Pediatric April 1988. I
Department
Radiology,
San
of the
Diego,
CA,
35209. 3
Department
of Pediatrics, University of Cincin-
of Medicine,
Cincinnati,
OH 45229-
4 Present address: Department of Radiology, Le Bonheur Children’s Medical Center, 848 Adams Ave., Memphis, TN 381 03. Address reprint requests to A. A. Kaufman.
S
University
of Cincinnati
College
Cincinnati, OH 45267. 0361 -803X/90/1 © American
543-0569
Roentgen
occurs
significance uncertain.
1990
frequently
in children
after blunt abdominal
trauma,
but the
of this finding and its implication for further diagnostic imaging This study addresses the following questions that have been
raised in the recent medical literature: (1) does the amount of hematuria predict the presence and extent of renal injury [1 2]; (2) is microhematuria with shock a better ,
of Radiology, University of Cincin-
nati College of Medicine, and Children’s Hospital Medical Center, Cincinnati, OH 45229-2899. 2 Present address: Radiology Associates of Birmingham, 1920 Huntington Rd., Birmingham, AL
nati College 2899.
clinical remain
March
Ray Society
of Medicine,
predictor of renal injury than microhematuria without shock [3, 4]; (3) would significant renal injuries be missed if imaging evaluation was not performed for asymptomatic patients with small amounts of hematuria [4-7]; (4) are patients with
head injury more likely to have hematuria
than those without
such injuries
[2, 6];
and (5) are significant renal injuries ever clinically occult in children [2]? We reviewed the medical records and abdominal CT scans of 256 children with blunt abdominal trauma and compared the clinical, laboratory, and imaging data of children who had renal injury with the data of those who did not.
Materials
and
Methods
Between August i98i and June i987, suspected multiorgan injury were examined Center
these
in Cincinnati.
patients,
The
medical
records
275 children by abdominal and
who are the basis of this report.
with blunt abdominal trauma and CT at Children’s Hospital Medical
CT examinations
could
be retrieved
in 256
of
STALKER
570
Children with blunt abdominal trauma were imaged by CT if single multiple organ injury was clinically suspected on the basis of physical findings, laboratory evaluation, or the circumstances of injury. All patients who were studied were judged to be hemodynamically stable after appropriate fluid and pharmacologic resuscita-
ET AL.
TABLE
AJA:154,
1: Hematuria
vs Renal
March
1990
Injury
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 06/01/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
or
emergency department. Examinations were performed either on a GE 8800 or a GE 9800 scanner; oral contrast medium tion
in the
and
IV contrast
medium
was
omitted
studies were performed by using dynamic scanning with rapid table incrementation. Details of our criteria for selecting patients and the examination technique have been published [8, 9]. Children with pelvic injuries and fractures were included in the analysis if abdominal injury was suspected clinically. For each patient, we recorded clinical and laboratory data including the microscopic and dipstick urinalyses, systolic and diastolic blood
pressure, hematocrit, symptoms, physical findings, Foley catheter placement and time of placement, and complications of renal injuries. The actual laboratory report of the amount of hematuria was recorded, but only the upper limit of the reported range of RBCs per high-powerfield
(RBC/HPF)
was used in data analysis.
hematuria
defined
more
systolic
and diastolic
were obtained accident.
as
blood cases,
were
For this study, [iO]. The initial recorded, whether they
department
or at the scene of the
than
pressures
in the emergency
In some
fluid
or
5 RBC/HPF
pharmacologic
resuscitation
had
been performed
before the first recording of the blood pressure. For purposes of analysis, hypotension was defined as a systolic or diastolic pressure more than two standard deviations below the mean forthat age [ii]. Renal injuries were graded I-V according to the scheme of Karp et al. [2]. Grade I injury included a small parenchymal injury without subcapsular or perirenal fluid, an uninjured anomaly. Grade II injury included an incomplete renal laceration, a small amount of subcapsular or perirenal fluid. Grade Ill injury represented extensive laceration or fracture, large perirenal fluid collection. Grade IV was a shattered kidney, multiple fragments. Grade V represented any vascular injury. For the purpose of this study, significant renal injury was defined as grade III or greater. In addition, we recorded the following: side of trauma,
precise
location
of injury
(upper
third,
middle
third,
lower third; central or peripheral), percentage volume of injured renal tissue, the presence and volume of perirenal fluid, and the presence of
intraperitoneal
associated
No Renal Injury (%)
Patients
Present Absent
31 (1 2) 4 (2)
75 (29) i 46 (57)
1 06 i 50
Totals
35 (14)
221 (86)
256 (100)
(%) (4i) (59)
given as a bolus were used. Oral contrast when the clinical situation dictated so. CT
medium
was
Renal Injury (%)
Hematuria
and
extraperitoneal
injuries
preexisting lesions or minor atraumatic abnormalities, we counted them as injuries. A direct relationship was seen between the amount of hematuria and the severity of renal injury (Fig. i). The mean number of RBCs on microscopic urinalysis increased as the
grade of renal injury increased.
In children
who had little or
no renal injury, RBC/HPF varied from 0 to more than 99. In children with more severe grades of renal injury ( grade Ill), the amount of hematuria was at least 99 RBC/HPF in all patients but one, a child with a renal pedicle injury who presented in shock but without hematuria. Thirty-eight patients had either systolic or diastolic hypotension at presentation. Six of these (1 6%) had imageable renal injuries, two of which were grade III or greater. Sixty patients had craniofacial injury. None of these had renal injuries without significant laboratory or physical abnormalities. This also was true in children whose abdomens were thought to be unassessable because of their depressed state of consciousness or because of previous IV administration of pancuronium bromide. Patients with head injuries had the same prevalence of hematuria (45#{176}Io) and renal injury (1 2%) as those without head injuries (41 % and 14%, respectively) (p > .05).
Discussion
In our study group, the amount of hematuria predicted both the presence and severity of renal injury. We found that the
including
head injuries. 1o0
Results
90
Of the 275 consecutive patients examined during the study period, medical records and imaging studies were available
80
for 256. There
70
were
1 70 boys
(66%)
and 86 girls (34%).
age range was 1-1 7 years (mean, 7.4 years; standard
The
devia-
(1 4%) had renal injury that could be diagnosed by using CT. The distribution of injuries was as follows: 1 5 patients had grade I injuries, 1 1 had grade II Thirty-five
patients
injuries, five had grade Ill injuries, one had grade IV injury, and three patients had grade V injuries. Nine patients had injuries of grade Ill or greater. One hundred six children (41 %) presented with hematuria (Table 1). Of these, 31 (29%) had renal injury on CT. Of the
150 children
who
presented
without
hematuria,
146 had
normal kidneys, and four had renal injuries (one renal pedicle injury; three small, nondescript, parenchymal hypodensities).
Although
60 I
tion, 4.1 years).
these hypodensities
may have represented
small,
3
50 40 30 20 10 0 3 Renal Inlury Grade
Fig. 1.-Graph shows direct relationship between amount of hematuria and severity of renal injury. Microscopic hematuria increased as grade of renal injury increased. RBC/HPF = RBCs per high-power field.
AJR:154,
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 06/01/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
larger
March
HEMATURIA
1990
the number
of RBCs
probability of significant to the work of Karp
predicted indicated,
in the urine,
BLUNT
the greater
the
V renal injury
who
presented
These data show that significant not have been missed
without
renal injuries
if CT examinations
hema-
with
studies
grade III) had not been
by Guice
et al. [5] and Fortune
et al.
[6], who found no significant renal injuries in patients with fewer than 30 RBC/HPF. Lieu et al. [1 2] examined 78 children who had excretory urography for blunt abdominal trauma and found no injuries in patients with less than 20 RBC/HPF and significant renal injury (grade III or greater) only if there was gross hematuria on too many ABCs to count on microscopic
analysis. On the basis ofthis study, the authors recommended that excretory urography (or CT) be performed if the number of RBCs is too numerous to count on microscopic analysis; if it is deemed
necessary
to diagnose
contusion,
that investi-
gation be undertaken when hematunia exceeds 20 RBC/I-IPF; and when there are associated clinical findings that suggest significant injury, urography also may be warranted. All of these studies, and our own, support the thesis that there are insignificant
levels
of hematunia
in children
after
blunt
trauma
that do not require further investigation. In our population, microhematuria with hypotension was not a more sensitive predictor of renal injury than was microhematunia without hypotension. The combination of these findings predicted only one of nine significant renal injuries, whereas hematuria without hypotension correctly predicted seven of nine. Nonetheless, hypotension was an important finding,
as the presence
of either
hematunia
or hypotension
or both correctly predicted all nine significant renal injuries (Table 2). An association between head trauma and hematunia has been noted [2, 6] before, and the mere presence of head trauma has been cited as sufficient indication for abdominal CT [1 3]. We found the same prevalence of hematuria in patients
with
head
TABLE
injuries
as in those
without
571.
TRAUMA
2: Renal
Injury Predicted
by Hematuna
or Hypotension Either
Injury Grade ?ill(n=9)
569
The Significance of Hematuria Children After Blunt Abdominal Trauma
H. Philip
Robert
Stalker1’2
A. Kaufman1’3’4 Kurt Stedje5
The clinical significance of hematuria in children who sustain blunt continues to be debated, as do the criteria for diagnostic imaging Previous reports have discussed the usefulness of certain clinical injury, such as the amount of hematuna present, the presence of injury,
and
the
presence
or absence
of symptoms
or findings
in
abdominal trauma in this population. predictors of renal shock or of head
on physical
examination.
To assess the value of such predictors of renal injury in children with posttraumatic hematuria, we reviewed and analyzed the medical records and abdominal CT examinations of 256 children with blunt abdominal trauma. One hundred six children (41%) had hematuria. Thirty-five patients (14%) had renal injury that could be diagnosed by using CT. Nine of these had clinically significant injuries according to our criteria. We found a direct relationship between the amount of hematuria and the severity of renal injury. Hypotension at presentation occurred in 38 patients and was an insensitive predictor of renal injury. The combination of hypotension and hematuna was no more sensitive than hematuria alone in predicting renal injury. Sixty patients had concomitant craniofacial injuries. This subgroup had the same prevalence of hematuria and renal injury as the group that did not have head injuries. There were no clinically occult renal injuries in the study population. Furthermore, we found that no normotensive child with fewer than 50 RBCs per high-power field had a significant renal injury, and conversely, all children with significant renal injuries had either large amounts of hematuna or shock.
154:569-571,
AJR
Hematuria Received August 14, 1989; accepted after revision October
24, 1989.
Presented
at the 31 st annual meeting
Society for Pediatric April 1988. I
Department
Radiology,
San
of the
Diego,
CA,
35209. 3
Department
of Pediatrics, University of Cincin-
of Medicine,
Cincinnati,
OH 45229-
4 Present address: Department of Radiology, Le Bonheur Children’s Medical Center, 848 Adams Ave., Memphis, TN 381 03. Address reprint requests to A. A. Kaufman.
S
University
of Cincinnati
College
Cincinnati, OH 45267. 0361 -803X/90/1 © American
543-0569
Roentgen
occurs
significance uncertain.
1990
frequently
in children
after blunt abdominal
trauma,
but the
of this finding and its implication for further diagnostic imaging This study addresses the following questions that have been
raised in the recent medical literature: (1) does the amount of hematuria predict the presence and extent of renal injury [1 2]; (2) is microhematuria with shock a better ,
of Radiology, University of Cincin-
nati College of Medicine, and Children’s Hospital Medical Center, Cincinnati, OH 45229-2899. 2 Present address: Radiology Associates of Birmingham, 1920 Huntington Rd., Birmingham, AL
nati College 2899.
clinical remain
March
Ray Society
of Medicine,
predictor of renal injury than microhematuria without shock [3, 4]; (3) would significant renal injuries be missed if imaging evaluation was not performed for asymptomatic patients with small amounts of hematuria [4-7]; (4) are patients with
head injury more likely to have hematuria
than those without
such injuries
[2, 6];
and (5) are significant renal injuries ever clinically occult in children [2]? We reviewed the medical records and abdominal CT scans of 256 children with blunt abdominal trauma and compared the clinical, laboratory, and imaging data of children who had renal injury with the data of those who did not.
Materials
and
Methods
Between August i98i and June i987, suspected multiorgan injury were examined Center
these
in Cincinnati.
patients,
The
medical
records
275 children by abdominal and
who are the basis of this report.
with blunt abdominal trauma and CT at Children’s Hospital Medical
CT examinations
could
be retrieved
in 256
of
STALKER
570
Children with blunt abdominal trauma were imaged by CT if single multiple organ injury was clinically suspected on the basis of physical findings, laboratory evaluation, or the circumstances of injury. All patients who were studied were judged to be hemodynamically stable after appropriate fluid and pharmacologic resuscita-
ET AL.
TABLE
AJA:154,
1: Hematuria
vs Renal
March
1990
Injury
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 06/01/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
or
emergency department. Examinations were performed either on a GE 8800 or a GE 9800 scanner; oral contrast medium tion
in the
and
IV contrast
medium
was
omitted
studies were performed by using dynamic scanning with rapid table incrementation. Details of our criteria for selecting patients and the examination technique have been published [8, 9]. Children with pelvic injuries and fractures were included in the analysis if abdominal injury was suspected clinically. For each patient, we recorded clinical and laboratory data including the microscopic and dipstick urinalyses, systolic and diastolic blood
pressure, hematocrit, symptoms, physical findings, Foley catheter placement and time of placement, and complications of renal injuries. The actual laboratory report of the amount of hematuria was recorded, but only the upper limit of the reported range of RBCs per high-powerfield
(RBC/HPF)
was used in data analysis.
hematuria
defined
more
systolic
and diastolic
were obtained accident.
as
blood cases,
were
For this study, [iO]. The initial recorded, whether they
department
or at the scene of the
than
pressures
in the emergency
In some
fluid
or
5 RBC/HPF
pharmacologic
resuscitation
had
been performed
before the first recording of the blood pressure. For purposes of analysis, hypotension was defined as a systolic or diastolic pressure more than two standard deviations below the mean forthat age [ii]. Renal injuries were graded I-V according to the scheme of Karp et al. [2]. Grade I injury included a small parenchymal injury without subcapsular or perirenal fluid, an uninjured anomaly. Grade II injury included an incomplete renal laceration, a small amount of subcapsular or perirenal fluid. Grade Ill injury represented extensive laceration or fracture, large perirenal fluid collection. Grade IV was a shattered kidney, multiple fragments. Grade V represented any vascular injury. For the purpose of this study, significant renal injury was defined as grade III or greater. In addition, we recorded the following: side of trauma,
precise
location
of injury
(upper
third,
middle
third,
lower third; central or peripheral), percentage volume of injured renal tissue, the presence and volume of perirenal fluid, and the presence of
intraperitoneal
associated
No Renal Injury (%)
Patients
Present Absent
31 (1 2) 4 (2)
75 (29) i 46 (57)
1 06 i 50
Totals
35 (14)
221 (86)
256 (100)
(%) (4i) (59)
given as a bolus were used. Oral contrast when the clinical situation dictated so. CT
medium
was
Renal Injury (%)
Hematuria
and
extraperitoneal
injuries
preexisting lesions or minor atraumatic abnormalities, we counted them as injuries. A direct relationship was seen between the amount of hematuria and the severity of renal injury (Fig. i). The mean number of RBCs on microscopic urinalysis increased as the
grade of renal injury increased.
In children
who had little or
no renal injury, RBC/HPF varied from 0 to more than 99. In children with more severe grades of renal injury ( grade Ill), the amount of hematuria was at least 99 RBC/HPF in all patients but one, a child with a renal pedicle injury who presented in shock but without hematuria. Thirty-eight patients had either systolic or diastolic hypotension at presentation. Six of these (1 6%) had imageable renal injuries, two of which were grade III or greater. Sixty patients had craniofacial injury. None of these had renal injuries without significant laboratory or physical abnormalities. This also was true in children whose abdomens were thought to be unassessable because of their depressed state of consciousness or because of previous IV administration of pancuronium bromide. Patients with head injuries had the same prevalence of hematuria (45#{176}Io) and renal injury (1 2%) as those without head injuries (41 % and 14%, respectively) (p > .05).
Discussion
In our study group, the amount of hematuria predicted both the presence and severity of renal injury. We found that the
including
head injuries. 1o0
Results
90
Of the 275 consecutive patients examined during the study period, medical records and imaging studies were available
80
for 256. There
70
were
1 70 boys
(66%)
and 86 girls (34%).
age range was 1-1 7 years (mean, 7.4 years; standard
The
devia-
(1 4%) had renal injury that could be diagnosed by using CT. The distribution of injuries was as follows: 1 5 patients had grade I injuries, 1 1 had grade II Thirty-five
patients
injuries, five had grade Ill injuries, one had grade IV injury, and three patients had grade V injuries. Nine patients had injuries of grade Ill or greater. One hundred six children (41 %) presented with hematuria (Table 1). Of these, 31 (29%) had renal injury on CT. Of the
150 children
who
presented
without
hematuria,
146 had
normal kidneys, and four had renal injuries (one renal pedicle injury; three small, nondescript, parenchymal hypodensities).
Although
60 I
tion, 4.1 years).
these hypodensities
may have represented
small,
3
50 40 30 20 10 0 3 Renal Inlury Grade
Fig. 1.-Graph shows direct relationship between amount of hematuria and severity of renal injury. Microscopic hematuria increased as grade of renal injury increased. RBC/HPF = RBCs per high-power field.
AJR:154,
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 06/01/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
larger
March
HEMATURIA
1990
the number
of RBCs
probability of significant to the work of Karp
predicted indicated,
in the urine,
BLUNT
the greater
the
V renal injury
who
presented
These data show that significant not have been missed
without
renal injuries
if CT examinations
hema-
with
studies
grade III) had not been
by Guice
et al. [5] and Fortune
et al.
[6], who found no significant renal injuries in patients with fewer than 30 RBC/HPF. Lieu et al. [1 2] examined 78 children who had excretory urography for blunt abdominal trauma and found no injuries in patients with less than 20 RBC/HPF and significant renal injury (grade III or greater) only if there was gross hematuria on too many ABCs to count on microscopic
analysis. On the basis ofthis study, the authors recommended that excretory urography (or CT) be performed if the number of RBCs is too numerous to count on microscopic analysis; if it is deemed
necessary
to diagnose
contusion,
that investi-
gation be undertaken when hematunia exceeds 20 RBC/I-IPF; and when there are associated clinical findings that suggest significant injury, urography also may be warranted. All of these studies, and our own, support the thesis that there are insignificant
levels
of hematunia
in children
after
blunt
trauma
that do not require further investigation. In our population, microhematuria with hypotension was not a more sensitive predictor of renal injury than was microhematunia without hypotension. The combination of these findings predicted only one of nine significant renal injuries, whereas hematuria without hypotension correctly predicted seven of nine. Nonetheless, hypotension was an important finding,
as the presence
of either
hematunia
or hypotension
or both correctly predicted all nine significant renal injuries (Table 2). An association between head trauma and hematunia has been noted [2, 6] before, and the mere presence of head trauma has been cited as sufficient indication for abdominal CT [1 3]. We found the same prevalence of hematuria in patients
with
head
TABLE
injuries
as in those
without
571.
TRAUMA
2: Renal
Injury Predicted
by Hematuna
or Hypotension Either
Injury Grade ?ill(n=9)
