Does A Positive Abdominal Examination In Blunt Trauma Patients Correlate To Positive Findings On CT?

Date First Published:

September 7, 2022

Last Updated:

September 14, 2022

Report by:

Briony Seden, Intercalated Medical Student, University of Plymouth (Plymouth University)

Search checked by:

Dr Laura Cottey, Plymouth University

Three-Part Question:

In [alert adult patients] does [positive findings on abdominal examination] correlate with [significant intra-abdominal findings on CT]

Clinical Scenario:

A 28 year old female is brought to the Emergency Department by ambulance after being in a road traffic accident. She is GCS 15 and on examination she is complaining of abdominal pain on palpation and has an abdominal seatbelt sign. You want to know how likely it is that a significant intra-abdominal injury (IAI) finding will be found on CT.

Search Strategy:

NICE Healthcare databases (1985-07/06/2022) including: AMED, PubMED, BNI, EMBASE, HBE, HMIC, Medline, PsycINFO, CINAHL

Search terms: Trauma, injury, abdominal, abdomen, examin* computed tomography, CT ti*ab

Outcome:

824 papers returned, 503 papers selected by title, 104 reviewed by abstract of which 79 were irrelevant. 25 papers were relevant and of sufficient quality and included in this BET. 12 papers were duplicated.

Relevant Paper(s):

Study Title	Patient Group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Does this adult patient have a blunt intra-abdominal injury? Nishijima DK, et al. 2012 USA	12 studies included for analysis. Total patient group of 10,757. Inclusion criteria; adult patients with any blunt abdominal trauma except for 2 studies which included only adult patients in motor vehicle collisions	Systematic Review	The presence or absence of abdominal tenderness does not include or exclude intra-abdominal injury. Rebound tenderness, abdominal distention, guarding, seat belt sign, and hypotension (systolic blood pressure <90 mm Hg) have likelihood ratios indicating the need for evaluation.	Sensitivity, % (95% CI) = 50 (35-65) Specificity, % (95% CI) = 91-95 Positive Likelihood Ratio (95% CI) = 5.6-9.9 Negative Likelihood Ratio (95% CI) = 0.53-0.55	Lack of physician’s clinical impression regarding the risk of IAI. Different inclusion/exclusion criteria used in each study. Old studies from more than 10 years ago.
			Seat Belt Sign	Sensitivity, % (95% CI) = 5 (0-10) Specificity, % (95% CI) = 99 (99-100) Positive Likelihood Ratio (95% CI) = 6.5 (1.8-24) Negative Likelihood Ratio (95% CI) = 0.96 (0.91-1.0)
			Rebound Tenderness	Sensitivity, % (95% CI) = 13 (6-20) Specificity, % (95% CI) = 97 (95-98) Positive Likelihood Ratio (95% CI) = 3.8 (1.9-7.6) Negative Likelihood Ratio (95% CI) = 0.90 (0.83-0.98)
			Abdominal Distention	Sensitivity, % (95% CI) = 26 (16-35) Specificity, % (95% CI) = 93 (91-95) Positive Likelihood Ratio (95% CI) = 3.7 (2.3-5.9) Negative Likelihood Ratio (95% CI) = 0.80 (0.70-0.91)
			Guarding	Sensitivity, % (95% CI) = 70 (57-81) Specificity, % (95% CI) = 57 (51-63) Positive Likelihood Ratio (95% CI) = 1.6 (1.3-2.0) Negative Likelihood Ratio (95% CI) = 0.52 (0.34-0.79)
			Abdominal pain as a symptom	Sensitivity, % (95% CI) = 71 (57-82) Specificity, % (95% CI) = 50 (44-57) Positive Likelihood Ratio (95% CI) = 1.4 (1.3-1.5) Negative Likelihood Ratio (95% CI) = 0.61 (0.46-0.80)
			Abdominal tenderness to palpation
How useful are clinical details in blunt trauma referrals for computed tomography of the abdomen? Beviss-Challinor KB et al. 2020 South Africa	139 patients who had undergone blunt trauma from 01 January 2018 - 31 March 2018 and had a CT scan. The clinical details on the request for the scan and the interpretation of the scan were compared.	Retrospective Study	Percentage of patients with a sign of intra-abdominal injury on CT following:	35%, p=0.05	The retrospective design meant that parameters were not specified in every case, thereby preventing determination of their diagnostic sensitivities and specificities. Only a univariate analysis was conducted so confounding variables were not controlled. The capture of data pertaining to imaging tests performed prior to CT was limited which made determining how the diagnosis was made difficult.
			Positive abdominal examination	0% p=0.05
			Negative abdominal examination
Correlating abdominal pain and intra-abdominal injury in patients with blunt abdominal trauma. Neeki M, et al. 2017 USA	594 patients brought to the ED following blunt trauma and assessed for abdominal pain on examination, then findings correlated with CT	Retrospective Study	Abdominal tenderness on physical examination	214	Due to its retrospective nature, this study was dependent on the varying documentation of the providers’ documentation and clinical acumen. Further, variation in provider documentation may also have impacted the identification and exclusion of patients with distracting injuries. Only abdominal tenderness was assessed, not other objective abdominal. Interpretation of the results are limited to the context of abdominal tenderness findings in correlation with CT imaging. The exclusion of patients with BAT who did not receive an abdominal or pelvic CT may have resulted in the missed inclusion of relevant cases.
		Retrospective Study	Positive CT findings from patients with abdominal tenderness on physical examination	78 (36.5%), p=<0.001
Patients with Abrasion or Ecchymosis Seat Belt Sign Have High Risk for Abdominal Injury, but Initial Computed Tomography is 100% Sensitive Shreffler J, et al. 2020 USA	425 patients presenting to ED with an abdominal seatbelt sign. Compared to CT findings of intra-abdominal injury.	Retrospective Study	Percentage of positive CT findings from patients with the following findings on abdominal examination:	45.4% p=<0.05	Shreffler J, et al. 2020 USA t425 patients presenting to ED with an abdominal seatbelt sign. Compared to CT findings of intra-abdominal injury.tRetrospective studytPercentage of positive CT findings from patients with the following findings on abdominal examination:tMany charts documented “seat belt sign” without further description. Any cases with uncertain type of seat belt sign were analysed as ‘unknown’, excluded from the abrasion/ecchymosis cohort analysis. Due to retrospective design and potential for missed subjects, an IAI rate of 36.1% may not predict the incidence of IAI in other populations. The ED cares for trauma patients in urban and rural areas with significant mechanism of trauma, thus, patients have a high probability of abdominal injuries. The study may have missed patients who presented to outlying facilities on return visits and were not transferred to the trauma centre for care.
			Abrasion	32.8% p=<0.05
			Ecchymosis	37.1% p=<0.05
			Abrasion and ecchymosis	53.6% p=<0.05
			Overall positive CT findings from a positive abdominal examination	36.1% p=<0.05
			Incidence of intra-abdominal injury (IAI) in the above sample group
Correlation Between Traumatic Skin and Subcutaneous Injuries and the Severity of Polytrauma Injury Klempka A, et al. 2021 Germany	30 patients presenting with blunt trauma who underwent CT scan following signs of superficial injury to the abdomen	Retrospective Study	Correlation between superficial injury of the abdominal cavity and an internal injury	Superficial injury (n=30) p=0.117	Very small sample size Retrospective observation of the superficial lesions on CT scans was not correlated with a clinical examination. The data were obtained only from one centre. Thus, the results should be confirmed by multi-centre studies on larger populations.
			Whole body CT	n = 11 %36.7
			Internal injury of abdomen	n = 19 %63.3
			No internal injury of abdomen
New scoring system for intra-abdominal injury diagnosis after blunt trauma Shojaee M, et al. 2014 Iran	261 patients who presented to ED following blunt trauma. Assessed for abdominal injury and then the findings compared to CT which was considered the gold standard.	Prospective Observational Study	CT identified 48 patients with IAI. The following signs were present at examination of the patient group.		The age range which mostly covers 21-30 years old misrepresents other age groups especially older adults and children. Investigators were not blinded to the purpose of this study.
New scoring system for intra-abdominal injury diagnosis after blunt trauma Shojaee M, et al. 2014 Iran	261 patients who presented to ED following blunt trauma. Assessed for abdominal injury and then the findings compared to CT which was considered the gold standard.	Prospective Observational Study	CT identified 48 patients with IAI. The following signs were present at examination of the patient group.	Percentage of patients = 62.5%tOdds ratio = 5.4 tLogistic regression = beta = 1.6tp = 0.05	The small assessed population is a potential limitation of the present study. A higher sample size may change other indices into statistically significant factors related to IAI diagnosis. The age range which mostly covers 21-30 years old misrepresents other age groups especially older adults and children. Investigators were not blinded to the purpose of this study.
			Abdominal pain	Percentage of patients = 10.4%tOdds ratio = 6.1 tLogistic regression = beta = 0.5tp = 0.8
			Abdominal guarding	Percentage of patients = 75%tOdds ratio = 10.0 tLogistic regression = beta = 2.9tp = 0.008
			Abdominal tnederness	Percentage of patients = 35.4%tOdds ratio = 2.0 tLogistic regression = beta = 0.2tp = 0.6
			Abdominal wall sign
Comparison of ultrasound and physical examination with computerized tomography in patients with blunt abdominal trauma Hekimoglu A, et al. 2019 Turkey	A total of 535 adult abdominal trauma patients: 359 males, 176 females, who underwent CT examination after positive physical examination	Retrospective Study	The ability of pain on abdominal palpation to predict presence of positive abdominal signs on CT	59%	Limitations of a retrospective study Abdominal examination was limited to palpation only, and no other injuries were included.
			Sensitivity	87%
			Specificity	70%
			Positive Estimated Value	81%
			Negative Estimated Value
A nomogram predicting the need for abdominal and pelvic computed tomography in blunt trauma patients: A retrospective cohort study. Lee JY, et al. 2017 South Korea	786 patients that were admitted to ED following blunt trauma. The study used their result to create a set of criteria for the requirement for CT.	Retrospective Study	Results of multivariate analyses of the need for abdominal and pelvic computed tomography	Odds ratio (95% confidence interval) = 1.036 (0.432–2.484)tP value = 0.937	Pain is a very subjective scoring system so there is significant variation between patients. There are limitations of a retrospective, single centre study.
			Laceration in torso region	Odds ratio (95% confidence interval) = 1.645 (0.294–9.210)tP value = 0.571
			Pain	Odds ratio (95% confidence interval) = 3.391 (2.135–5.386)tP value = <0.001
			Unevaluable	Odds ratio (95% confidence interval) = 1.284 (0.227–7.274)tP value = 0.778
			Positive	Odds ratio (95% confidence interval) = 7.351 (1.449–37.287)tP value = 0.016
			Peritoneal Signs
			Unevaluable
			Positive
Computed tomography and blunt abdominal injury: patient selection based on examination, haematocrit and haematuria. Richards JR, Derlet RW 1997 USA	444 patients evaluated by CT for the presence of intra-abdominal injury, based on multiple factors including physical examination	Retrospective Study	Ability of abdominal tenderness to screen for IAI.	63% (95% CI 48-77%)	Limitations of a retrospective review A selection bias existed in that ICU admissions and patients going directly to the operating room were omitted. The presence of abdominal tenderness was subjective, and the location, presence of rebound, guarding, or distention was not characterised.
			Sensitivity	65% (95% CI 60-70%)
			Specificity	19% (95% CI 13-25%)
			Positive Predictive Value	93% (95% CI 90-96%)
			Negative Predictive Value
Intra-abdominal injury following blunt trauma becomes clinically apparent within 9 hours Jones EL, et al. 2014 USA	285 patients identified from the Trauma Registry that were assessed in the ED for the length of time it took for signs of blunt abdominal trauma to appear. Presence of injury was then confirmed by CT	Retrospective Study	Patients who displayed positive abdominal signs confirmed by CT	82%	The retrospective identification of specific clinical variables prompting the imaging or intervention is prone to error. In addition, the hierarchy of clinical signs and symptoms was designed for ease of categorization based on the authors' experience but may not be applicable in all situations.
Selective use of computed tomography and diagnostic peritoneal lavage in blunt abdominal trauma Grieshop NA et al. 1995 USA	The records of 1096 patients that attended a Level 1 trauma centre following blunt trauma were reviewed for findings on initial physical examination, and results of CT, DPL, laparotomy and postmortem.	Retrospective Study	Percentage of patients with an abnormal abdominal examination that were found to have significant IAI	17.80%	Retrospective design means that correlation does not equal causation. This was only carried out at a single centre so the results may not be generalisable.
		Retrospective Study	The ability of an abnormal abdominal examination to predict IAI	Odds ratio = 10.3 p = 0.0001
Accuracy of clinical, laboratory, and computed tomography findings for identifying hollow viscus injury in blunt trauma patients with unexplained intraperitoneal free fluid without solid organ injury. Jost E, et al. 2017 Canada	39 patients from the Southern Alberta Trauma Registry who had a CT for blunt abdominal trauma	Cross-sectional Study	Association of categorically measured physical findings with likelihood of hollow viscus injury at presentation	Number with therapeutic odds ratio = 6tPositive likelihood ratio = approaches infinitytSensitivity = 37.5% (15.2–64.6)tSpecificity = 100% (85.2–100)	Imaging alone even with modern technologies cannot reliably predict hollow viscus injury A corresponding limitation is that the confidence intervals for the positive likelihood ratios for abdominal distention crossed unity, likely due to the small sample size of the study Another limitation is that the statistical methods do not suggest specific numerical values which increase the suspicion of hollow vicus injury. The aim of this study was not to confirm values that others had suggested, but rather to identify which imaging, physical, and laboratory findings were diagnostically accurate
			Peritonitis	Number with therapeutic odds ratio = 4tPositive likelihood ratio = approaches infinitytSensitivity = 25% (7.27–52.4)tSpecificity = 100% (85.2–100)
			Seatbelt Sign	Number with therapeutic odds ratio = 6tPositive likelihood ratio = 1.73 (0.634–4.69)tSensitivity = 37.5% (15.2–64.6)tSpecificity = 78.3% (56.3–92.5)
			Abdominal Distention
Identifying severe abdominal injuries during the initial assessment in blunt trauma patients. Farrath S, et al. 2013 Brazil	331 blunt trauma patients who were admitted to the ED and had a CT or laparotomy.	Retrospective Study	Comparison of qualitative variables between groups A (severe abdominal injury AIS >= 3) and B (abdominal injury AIS <3) as displayed on CT: univariate analysis. Data presented as percentage related to the presence of the variable in a particular group.	Group A n = 101 61.4% Group B n = 230 28.7% p <0.001	Patients with distracting injuries and lower GCS were included in the study The sample group was 80% male so the results may not be applicable to females.
		Retrospective Study	Altered Abdominal Examination
Computed tomography for blunt abdominal trauma in the ED: a prospective study. Richards JR, Derlet RW. 1998 USA	196 patients receiving a CT in the ED of a Level 1 trauma centre following blunt trauma	Prospective Study	Comparison of patients with and without IAI confirmed by CT	Without IAI (n = 174) 102 With IAI (n = 22) 18 p = 0.04 Odds ratio (95% CI) 3.2 (1.0,9.8)	Some of the variables requested on the original questionnaire required subjective evaluation from the examining physician, including the chest and abdominal examination, reason for obtaining the scan, and planned disposition. In addition, there were different levels of training of the examining physicians, from intern to attending. The definition of IAI included any abnormality likely caused by trauma. There were many nonoperative conditions identified, such as renal contusion.
			Abdominal Tenderness	82% (95% CI 60-95%)
			As a screen for intra-abdominal injury, confirmed by CT, abdominal examination was:	41% (34-49%)
			Sensitivity	15% (9-23%)
			Specificity	95% (87-99%)
			Positive Predictive Value
			Negative Predictive Value
Predictors for the selection of patients for abdominal CT after blunt trauma: a proposal for a diagnostic algorithm. Deunk J, et al. 2010 Netherlands	1040 patients admitted to a Level 1 ED who met the high-energy trauma protocol.	Observational Prospective Study	Crude OR With 95% CI for >1 Traumatic Injury on Abdominal CT, Using Univariate Logistic Regression Analysis	n = 248 OR = 3.60 CI = 2.67–4.86	Although CT is a very sensitive modality to detect abdominal injuries, it is known to be less than 100% sensitive. In this study, CT was false-negative in 0.9% of the patients, mostly consisting of hollow-visceral and pancreatic injuries. The data in this study were derived from a blunt trauma population in a single Llevel 1 trauma centre. The data does not necessarily reflect other less injured populations.
		Observational Prospective Study	Abnormal examination of the abdomen and/or pelvis
Admission or observation is not necessary after a negative abdominal computed tomographic scan in patients with suspected blunt abdominal trauma: results of a prospective, multi-institutional trial Livingston DH, et al. 1998 USA	2744 patients admitted to one of four Level 1 trauma centres following blunt trauma. Patients underwent serial abdominal examination and CT to assess injuries.	Prospective study	Percentage of patients with physical abdominal findings of bruising, tenderness or guarding	61%	Possible variation in documentation between hospitals and variation in practice.
		Prospective study	Percentage of patients with positive abdominal findings that had an abnormal CT scan	26%
Injuries distracting from intraabdominal injuries after blunt trauma Ferrera PC, et al. 1998 USA	350 patients admitted to a Level 1 ED following blunt trauma. Split into two groups depending on the presence or absence of abdominal pain and correlated with findings on CT.	Prospective study	The values of abdominal pain or tenderness to predict a positive finding of IAI on CT	82% (95% confidence interval [CI], 78% to 86%)	The study is limited in that patients either discharged from the ED or admitted to the hospital without receiving abdominal CT or DPL may have had missed IAI. There was no outpatient follow-up on these patients and it is possible that they were seen at other hospitals presenting with symptoms referrable to delayed diagnosis of IAI (eg. splenic ruptures or bowel perforations).
			Sensitivity	45%
			Specificity	21%
			Positive Predictive Value	93% (95% CI, 90% to 96%)
			Negative Predictive Value
Comparison of clinically suspected injuries with injuries detected at whole-body CT in suspected multi-trauma victims. Shannon L, et al. 2015 UK	588 multi-trauma patients were enrolled. Their CT request cards were used to discover the clinical suspicion of injury and then compared to the CT reports	Prospective study	Percentage of patients with suspicion of intra-abdominal injury on the CT request card which had IAI on CT	31%	Inability to analyse the initial findings that led the physician to suspect whether a body area was injured and how this compared to CT findings, there is limited data to indicate whether the suspicion was based on clinical findings, mechanism, or a combination.
Correlation Between Traumatic Skin and Subcutaneous Injuries and the Severity of Polytrauma Injury. Klempka A, et al. 2020 Germany	250 patients following blunt trauma, who were assessed in ED, need for CT identified and findings reviewed to see if matched assessment	Retrospective study	Correlation between superficial injury of the abdominal cavity and an internal injury	n=11 %=36.7	The assumption that all body areas would be accessible for clinical examination to the same extent as for cross-sectional imaging in polytraumatic patients. The retrospective observation of the superficial lesions on CT scans was not correlated with a clinical examination. Data was obtained only from one university hospital centre. Thus, observations should be confirmed by multi-centre studies on larger populations.
			Internal injury of the abdomen	n=19 %=63.3
			No internal injury of the abdomen
A negative computed tomography may be sufficient to safely discharge patients with abdominal seatbelt sign from the emergency department: A case series analysis Barmparas G, et al. 2018 USA	196 patients admitted following blunt trauma with an abdominal seatbelt sign to a Level 1 trauma centre, whose work up and CT results were examined	Retrospective study	Percentage of patients who had an abdominal seatbelt sign with a positive finding on CT n = 183	37.70%	Findings should be interpreted with caution given the retrospective nature of this work and the potential associated bias. There is a possibility that patients with an abdominal seat belt sign might not have been captured. Similarly, it is possible that the presence of a seatbelt sign was not documented in the chart. Findings might not apply to pregnant patients. In addition, seven patients did not undergo a CT in violation of the institutional protocol and although they did not re-present in a delayed fashion, the possibility of them presenting to another institution cannot be excluded.
Negative computed tomography can safely rule out clinically significant intra-abdominal injury in the asymptomatic patient after blunt trauma: Prospective evaluation of 1193 patients. Benjamin E, et al. 2018 USA	1193 blunt trauma patients admitted to a Level 1 trauma centre who had a CT within 24 hours, results were then compared to the work up characteristics	Prospective study	The presence of positive abdominal examination compared to CT findings	Percentage of study group: 21.2% (253 patients) Percentage with positive CT findings 22.1% (56 patients)	Once discharged, the majority of patients had limited follow-up and delayed presentation of injuries may have been missed. Although abdominal imaging is obtained liberally as an institutional practice, there was no algorithm guiding the decision of which patients underwent cross-sectional imaging. This potentially inflates the denominator of negative studies and leaves a potentially unstudied population of patients that may have had missed injury but were never imaged.
			Abdominal pain	Percentage of study group: 38.0% (453 patients) Percentage with positive CT findings 23.0% (104 patients)
			External signs of trauma
Blunt abdominal trauma patients: can organ injury be excluded without performing computed tomography? Poletti PA, et al. 2004 USA	714 patients admitted to a Level one trauma centre, where CT was ordered based on clinical and laboratory results	Prospective study	Multivariate analysis of abdominal examination to predict an IAI as defined by CT results	Level (%) = 9 Sensitivity (%)= 26 PPV (%) =32 Specificity (%) = 93 NPV (%) = 91	Only patients already selected to undergo CT were enrolled. The decision not to perform CT in a certain number of patients was made on the basis of criteria that could not be analysed. However, because most of the patients with suspicion of blunt abdominal trauma undergo abdominal CT, this limitation only concerns a very small number of cases and will not contribute any consistent bias to the results. Not all patients had data forms completed may have led to an unmeasured selection bias. A substantial amount of data was not available on the patient notes.
			Guarding	Level (%) = 33 Sensitivity (%)= 46 PPV (%) =16 Specificity (%) = 69 NPV (%) = 91
			Tenderness	Level (%) = 1 Sensitivity (%)= 5 PPV (%) =44 Specificity (%) = 99 NPV (%) = 89
			Rebound tenderness	Level (%) = 5 Sensitivity (%)= 13 PPV (%) =34 Specificity (%) = 97 NPV (%) = 89
			Distention	Level (%) = 48 Sensitivity (%)= 68 PPV (%) =17 Specificity (%) = 55 NPV (%) = 93
			GCS, guarding and tenderness
Intra-abdominal Injury Following Blunt Trauma: Identifying the High-Risk Patient Using Objective Risk Factors Mackersie RC, et al. 1989 USA	The records of 3223 major trauma patients admitted to the ED who sustained blunt injury were reviewed. For each patient, the presence or absence of a significant intra-abdominal injury (defined as an injury for which operative repair was required) was tabulated.	Prospective observational study	Correlation between abdominal examination and intra-abdominal injury n = 1648	35%	The study didn’t specifically analyse the factors of a positive abdominal examination and was more focussed on risk factors for blunt trauma associated abdominal injury.
			Abdominal Tenderness	0.40%
			Abdominal Distention
Prediction of blunt traumatic injury in high-acuity patients: bedside examination vs computed tomography. Smith CB, et al. 2011 USA	Convenience sample of 400 trauma patients that were assessed by 18 emergency medicine physicians who completed a data sheet for expected injuries on CT	Prospective study	Diagnostic positive abdominal examination for emergency physician bedside assessment compared with CT as the gold standard for the presence or absence of clinically significant injury	Prevalence (95% CI) =29.2 (24.4-34.5) % Sensitivity (95% CI) =18.8 (14.8-23.5) % Specificity (95% CI) =74.3 (69.2-78.9) % NPV (95% CI) =74.3 (69.2-78.9) % PPV (95% CI) =72.0 (66.8-76.7)	Convenience sampling could have led to selection bias in including sicker patients, as these are the ones physicians might have been most likely to remember to include in the study. This seems probable given the high number of intubated patients, with almost half intubated prehospital or immediately in the ED. The trauma department requests whole-body CT on all Level 1 patients with blunt trauma, with very few exceptions. This practice policy often leads to body regions that are scanned despite little initial clinical concern for injury instead of admitting to the hospital for serial examinations and close observation. Because the study evaluated 1 group of attending emergency medicine physicians, it is possible that other physician groups could have dissimilar prediction ability. Because this was an unstructured rating system, there may also be considerable heterogeneity among physicians as to what they considered “very low,” “low,” “intermediate,” and so on.
			Positive if pretest rating is: high	% Sensitivity (95% CI) =80.6 (75.8-84.6) % Specificity (95% CI) =80.6 (75.8-84.6) % NPV (95% CI) = 80.6 (75.8-84.6) % PPV (95% CI) = 58.5 (53.0-63.9)
			Positive if pretest rating is: intermediate	% Sensitivity (95% CI) =84.4 (79.9-88.0) % Specificity (95% CI) =50.6 (45.1-56.2) % NPV (95% CI) = 88.7 (84.7-91.8) % PPV (95% CI) = 41.3 (36.0-46.9)
			Positive if pretest rating is: low	% Sensitivity (95% CI) =97.9 (95.5-99.1) % Specificity (95% CI) =16.3 (12.6-20.9) % NPV (95% CI) = 95.0 (91.9-97.0) % PPV (95% CI) = 32.5 (27.5-37.9)
			Positive if pretest rating is: very low
. Prospective study of the clinical predictors of a positive abdominal computed tomography in blunt trauma patients Beck D, et al. 2004 USA	213 patients who presented to a Level I trauma centre as a trauma team alert and underwent an abdominal CT scan. The trauma team leader filled out a data sheet before the CT scan documenting the indications for CT	Prospective Study	Indications for CT	Positive CT (n = 56) = 25 (45%) Negative CT (N = 157) = 73 (46%) p-Value =0.81	Only small sample sizes used which limits generalisability
			Abnormal abdomen exam (tender/ distended)	Positive CT (n = 56) = 7 (12%) Negative CT (N = 157) = 14 (9%) p-Value =0.44
			Visible abdomen/ pelvis trauma

Author Commentary:

This systematic review aimed to ascertain if a positive abdominal examination in alert adults following blunt trauma was indicative of positive findings of IAI on CT. The literature on this subject area is plentiful, but variation between papers in methodology and results makes it hard to provide an accurate conclusion, with sensitivity ranging from 5% to 97.9%. It is also recognised that many studies had small sample sizes, making statistical calculations unreliable, and the results are rarely generalisable to other populations. Furthermore, the large number of retrospective studies means that data collection was limited, and detailed information on the abdominal examination was hard to obtain.

The presence or absence of abdominal tenderness does not include or exclude IAI; and distention, guarding, ASBS and rebound tenderness suggest the need for further evaluation. Pain appears to have the highest sensitivity and NPV across the studies, making it the sign most accurate in predicting CT findings. In contrast, an ASBS, ecchymosis and abrasion seem to be the least associated, possibly due to the superficial nature of these injuries. Therefore, it can be concluded that a positive abdominal examination does not significantly correlate with a positive CT finding, and hence is too unreliable to be depended upon in trauma work up protocols.

Bottom Line:

The clinical bottom line is that a positive abdominal examination does not conclusively correlate with a positive CT finding and therefore should be interpreted with caution.

References:

Nishijima DK, et al. . Does this adult patient have a blunt intra-abdominal injury?
Beviss-Challinor KB et al.. How useful are clinical details in blunt trauma referrals for computed tomography of the abdomen?
Neeki M, et al.. Correlating abdominal pain and intra-abdominal injury in patients with blunt abdominal trauma.
Shreffler J, et al.. Patients with Abrasion or Ecchymosis Seat Belt Sign Have High Risk for Abdominal Injury, but Initial Computed Tomography is 100% Sensitive
Klempka A, et al.. Correlation Between Traumatic Skin and Subcutaneous Injuries and the Severity of Polytrauma Injury
Shojaee M, et al. . New scoring system for intra-abdominal injury diagnosis after blunt trauma
Shojaee M, et al. . New scoring system for intra-abdominal injury diagnosis after blunt trauma
Hekimoglu A, et al. . Comparison of ultrasound and physical examination with computerized tomography in patients with blunt abdominal trauma
Lee JY, et al.. A nomogram predicting the need for abdominal and pelvic computed tomography in blunt trauma patients: A retrospective cohort study.
Richards JR, Derlet RW. Computed tomography and blunt abdominal injury: patient selection based on examination, haematocrit and haematuria.
Jones EL, et al.. Intra-abdominal injury following blunt trauma becomes clinically apparent within 9 hours
Grieshop NA et al.. Selective use of computed tomography and diagnostic peritoneal lavage in blunt abdominal trauma
Jost E, et al.. Accuracy of clinical, laboratory, and computed tomography findings for identifying hollow viscus injury in blunt trauma patients with unexplained intraperitoneal free fluid without solid organ injury.
Farrath S, et al.. Identifying severe abdominal injuries during the initial assessment in blunt trauma patients.
Richards JR, Derlet RW.. Computed tomography for blunt abdominal trauma in the ED: a prospective study.
Deunk J, et al.. Predictors for the selection of patients for abdominal CT after blunt trauma: a proposal for a diagnostic algorithm.
Livingston DH, et al.. Admission or observation is not necessary after a negative abdominal computed tomographic scan in patients with suspected blunt abdominal trauma: results of a prospective, multi-institutional trial
Ferrera PC, et al.. Injuries distracting from intraabdominal injuries after blunt trauma
Shannon L, et al.. Comparison of clinically suspected injuries with injuries detected at whole-body CT in suspected multi-trauma victims.
Klempka A, et al.. Correlation Between Traumatic Skin and Subcutaneous Injuries and the Severity of Polytrauma Injury.
Barmparas G, et al.. A negative computed tomography may be sufficient to safely discharge patients with abdominal seatbelt sign from the emergency department: A case series analysis
Benjamin E, et al.. Negative computed tomography can safely rule out clinically significant intra-abdominal injury in the asymptomatic patient after blunt trauma: Prospective evaluation of 1193 patients.
Poletti PA, et al.. Blunt abdominal trauma patients: can organ injury be excluded without performing computed tomography?
Mackersie RC, et al.. Intra-abdominal Injury Following Blunt Trauma: Identifying the High-Risk Patient Using Objective Risk Factors
Smith CB, et al.. Prediction of blunt traumatic injury in high-acuity patients: bedside examination vs computed tomography.
Beck D, et al.. . Prospective study of the clinical predictors of a positive abdominal computed tomography in blunt trauma patients