WTA: Significance of Contrast Blush on CT Scan for Splenic Injuries

WTA: Significance of Contrast Blush on CT Scan for Splenic Injuries

Purpose and Objectives

The spleen is the most commonly injured intra-abdominal solid organ following blunt injury in adults (1-3). Current guidelines for blunt splenic injury (BSI) treatment in adults favor non-operative management due to avoided operative morbidity and mortality as well as to averting the immunologic consequences of splenectomy (4). These guidelines are based primarily on retrospectively collected data and opinion. In the vacuum left by the lack of even a single randomized prospective multi-institutional trial examining this injury, several competing treatment protocols have emerged (5 – 10).

Much of the controversy surrounding BSI management in adults has to do with management of vascular abnormalities observed on computed tomography of the abdomen near to the time of injury. These abnormalities can take the form of intra-parenchymal contrast accumulation (which will be defined for the purposes of this protocol as splenic pseudoaneurysm), or extravasation of contrast from the spleen not confined to the parenchyma (which will be defined for the purposes of this protocol as a contrast blush). The management of both abnormalities is controversial (5-10). Immediate splenectomy, splenic angiography with embolization, and observation only are some of the treatment strategies used to manage these injuries. However, what is not known is the natural history and splenic salvage rates using these contrasting management techniques. Thus, there is an urgent need for a series of multi-institutional trials (including observational and randomized trials) to help determine the most advantageous BSI management strategy. Until these trials are completed, adoption of guidelines for BSI management will continue to be variable and injured patients will continue to receive, potentially, sub-optimal treatment.

To fill the significant knowledge gapwe propose collecting data from adult patients with BSI from multiple institutions across the United States and pursuing the following specific aims:

Ascertain the in-hospital risk of splenectomy for adult patients with BSI who present with or develop a splenic pseudoaneurysm, contrast blush or both while in-hospital.

Determine factors related to failure of non-operatively managed blunt splenic injuries with splenic pseudoaneurysm or contrast blush or both.

Accomplishing Specific Aim 1 will provide a description of the natural history of splenic pseudoaneurysm and contrast blush using retrospectivelycollected data. This information is critical so that adequately powered clinical trials for diagnosis, treatment, and follow-up care involving patients with BSI can be designed. By utilizing detailed information about each subject we will be able to address Specific Aim 2. We will determine the physiologic, injury and management factors associated with failure of non-operative management of BSI in an effort to define patients that may benefit from early splenectomy.

Background and significance

Of the millions of adults in the United States who are injured every year, nearly 39,000 will be admitted with a BSI (1-3). Over the last two decades there has been a shift in management of adults with BSI from primarily operative to primarily non-operative. While attempted non-operative management is thought to be the best strategy for adults who are hemodynamically stable with BSI, there are still big questions regarding the optimal management of these patients due to a lack of prospectively collected data. Even the most recently published guidelines for the management of adults with BSI are based on retrospective reviews and expert opinion (4). Basic information such as the rate of splenectomy after attempted non-operative management has not been prospectively collected. The retrospective data that do exist indicate that a little over 10% of patients with non-operatively managed BSI suffer early failure resulting in splenectomy within 5 days of injury (11-13).

Factors associated with failed non-operative management of BSI are also controversial in the extant literature. It may be that early failures of nonoperative management represent a group of patients who were inappropriately observed. Age greater than 55, increasing grade of spleen injury, presence of moderate to large hemoperitoneum on imaging studies, periods of hypotension and tachycardia, as well as the presence of splenic pseudoaneurysm or contrast blush have been suggested as potential risk factors associated with splenectomy after attempted non-operative management (12,16).

One controversial question in the management of BSI has to do with the management of patients with splenic pseudoaneurysm or contrast blush. Some advocate early splenic angiography and embolization (5,6). Using this protocol successful splenic salvage rates have been reported as 92% (6). Another strategy utilized in an attempt to maximize splenic salvage is to screen for splenic pseudoaneurysms within the first 24 – 48 hours of non-operative management of BSI. Using a splenic pseudoaneurysm screening protocol, Davis and colleagues reported an improvement in non-operative success in the setting of BSI to 94% (7). On the other hand, some groups advocate for early splenectomy in the setting of splenic pseudoaneurysm or contrast blush (9,17).

Thelong-term goalof this line of research is to advance the care of injured patients with BSI so that splenic salvage is maximized while morbidity and mortality are minimized. Achieving the specific aims of this proposal will be a significant contribution because it will be a step in a continuum of research that is expected to lead to the development of management strategies that will result in subjecting adults with BSI to the least risk while preserving the most spleens.

Research Plan / Methods

Study Design

This study will be a retrospective data collection study. Patients will be screened for eligibility based on inclusion/exclusion criteria when they are admitted. Data collection from a review of the patient chart will include patient demographics, past medical history (including calculation of the Charlson Comorbidity Index) injury related details, physiologic details, radiographic studies and any spleen related interventions that took place prior to study enrollment. Overall, we expect that each patient will be managed according to the local standard of care at each participating institution. The study protocol will be only to collect data obtained during standard of care treatment given to each subject as well as the outcome from the treatment. Thus, there is minimal overall study related risk to participating subjects.

Study Population

The study sample will be derived from patients admitted to the participating centers. The inclusion criteria consist of adult (age ≥ 18) persons who suffer a BSI and who have either a splenic pseudoaneurysm or contrast blush on computed tomography of the abdomen with contrast within 24 hours of admission. Persons will be excluded from the study if they: 1) are more than twenty-four hours from the time of injury when they are admitted to the hospital; 2) have a history of a previous splenic injury; 3) have a history of surgery involving the spleen; 4) have a history of a significant bleeding disorder (eg. Factor VII deficiency, Factor VIII deficiency); 5) are pregnant women (assessed by a urine pregnancy test); 6) or who have a history of any of the following:

Hereditary elliptocytosis
Hereditary spherocytosis
sickle cell disease
Thallassemia
Hodgkins or Non-Hodgkins lymphoma
Other lymphomas
Leukemia
Polycythemia vera
Myeolofibrosis
Metabolic storage diseases
Amyloidosis
Splenic vein thrombosis
Cirrhosis
Splenic cysts
Sarcoidosis
Systemic lupus erythematosous

We plan to make recruitment of women and minorities a priority for this project and we will include persons of any race who are eligible for the study. Persons under the age of 18 will not be included because the management of children with BSI is much less controversial and because children are much less likely to suffer delayed splenectomy after BSI due to the anatomy of the spleen in a child.

Recruitment and Consent Procedures

The institutional PI or his or her designee will carry out screeningfor this study. At each study site, similar screeningprocedures will be followed. If the patient meets inclusion criteria and does not meet the exclusion criteria, the patient chart will be reviewed for required data points.

Study data collected for this study is extracted from the patient chart and therefore no data will be obtained outside of data collected during standard of care procedures.

Randomization Procedures

Subjects will not be randomized in this study.

List procedures specific to the subjects being studied

Participating subjects will not undergo any procedures associated with this study except to obtain data regarding treatment. Each institution is expected to maintain treatment of subjects in accordance with the local standard of care. Below is a list of study related data that will be obtained at various time-points.

A. At study enrollment

1. Demographic data and past medical history including Charlson Comorbidity Index and use of anti-platelet drugs and/or warfarin as an outpatient will be recorded. Smoking history and past surgical history will also be obtained.

2. Admission physiologic characteristics including admission vital signs and pertinent laboratory data.

3. Injury characteristics including Abbreviated Injury Scale for each system, the Injury Severity Score, Glasgow Coma Score, spinal cord injury level (if the patient has a spinal cord injury) and any associated intra-abdominal injuries.

4. Spleen injury characteristics including AAST injury grade. Admission computed tomography scans will be uploaded to the data management system.

B. At the time of any spleen related intervention (in-patient)

1. Splenic angiography and/or embolization

2. Splenorrahphy

3. Splenectomy

For each of these interventions, details regarding vital signs and reason for the intervention will be obtained. If angioembolization was performed, details regarding the type of embolization material used and the location of embolization (main splenic artery vs selective embolization) will be obtained. Pertinent imaging will also be uploaded to the data management system.

C. At Hospital Discharge

1. Discharge related information including length of stay, length of intensive care unit stay and discharge disposition.

Details about the Charlson Comorbidity Index:

Charlson Comorbidity Index (CCI): Since injured persons may present to the trauma center with pre-existing medical conditions that could impact quality of life and functional capacity as well as survival, we will adjust for pre-existing medical disease using the CCI. The CCI contains 19 categories of comorbidity, which are primarily defined using ICD-9 diagnoses codes. Each category has an associated weight, taken from the original Charlson paper (18), which is based on the adjusted risk of one-year mortality. The overall comorbidity score reflects the cumulative increased likelihood of one-year mortality. The CCI has been utilized to adjust for comorbidities in studies of injured persons and can be given in a self report form (19,20). The CCI will be calculated at the time of study enrollment.

Benefits to subjects

This study does not change the way patients are treated, thus, there is no direct benefit associated with study participation. However, there is the possibility from the data gathered in this study that future patients with BSI may benefit.

Risks to subject and/or fetus

Note: For minimal risk studies (such as questionnaires/surveys) where losses of confidentiality or psychological stress are the only risks, these need to be listed. For example: There are no physical risks associated with this study because there are no medications or treatments being given to subjects for this study. There is, however, the potential risk of loss of confidentiality. Every effort will be made to keep each subject’s information confidential; however, this cannot be guaranteed.

Drug/Device Information

There are no devices used in this study.

Safety

Since there are no treatments prescribed in the protocol there is minimal risk to subjects. Subject confidentiality is paramount in this study since we will be obtaining protected health information. The details of how study related data are to be handled are outlined in the confidentiality section below.

Data Integrity/ Data Management / Monitoring Plan

Data will be maintained in an online, secure, HIPAA-compliant database administered by the American Association for the Surgery of Trauma. The website is password protected. Each site PI will have access to subjects enrolled from their site only. No PI will have access to the entire database. The study PI, however, will have the ability to monitor data input from each participating site. At least monthly, the number of subjects enrolled will be determined and the amount of missing data will be determined by the study PI. To ensure consistency among sites regarding data entry, a study related data dictionary will be developed. These efforts will help to ensure that the data gathered as part of this study is of the highest quality possible.

DSMB

Since this is a retrospective data collection study with no study related interventions, there is no need for a DSMB.

Alternative treatments

Subjects will receive treatment in accordance with the local standard of care for their injuries, including the injury to the spleen, regardless of study participation status.

Confidentiality

All paper research records will be stored in locked file cabinets and will be accessible only to research personnel. All electronic research records will be computer password protected and also accessible only to research personnel. Research records will be entered into a computerized study database and will be labeled with a code. A master key which links the subjects’ names with the codes for each research record will be maintained at the local investigative site until the study has ended and all the information has been collected and verified with the hospital chart.

PHI collected in this study may include medical history, the results of physical exams, lab tests, x-ray exams, and other diagnostic and treatment procedures, as well as basic demographic information. Institutional Review Boards (IRB) may review PHI as part of their responsibility to protect the rights and welfare of research subjects. PHI will not be used or disclosed to any other person or entity, except as required by law, or for authorized oversight of this research study by other regulatory agencies, or for other research for which the use and disclosure of your PHI has been approved by the IRB. PHI will be used until the study is completed.

No HIPAA will be collected as part of the data set for this study. Subject data sets will be assigned a unique number. Subjects will not be identified in any presentations or publications based on the results of this research study.

Compensation

Subjects will not be compensated for participation in this study.

Costs to Subjects

No additional costs will be incurred by a subject who participates in this study.

Plans for data analysis, including justification of sample size
Descriptive and Preliminary Analyses: Means, standard deviations and frequencies (where appropriate) will be calculated for the entire cohort for demographic data, the CCI, as well as admission ISS, and baseline injury characteristics.

Specific Aim #1: Ascertain the in-hospital risk of splenectomy for adult patients with BSI who present with or develop a splenic pseudoaneurysm, contrast blush or both while in-hospital.

Main Outcome: Splenectomy at anytime after study enrollment up to discharge.

Analysis Method: Splenectomy rates will be calculated at various time points after injury. The numerator for each calculation will be the person time contributed by the patients who have had a splenectomy by the time point. The denominator will be person-time at risk for the cohort up to that time point. The risk of splenectomy at each time-point will also be calculated. For these calculations the numerator will be the number of patients who have had a splenectomy up to the time point of interest. The denominator will be the total number of persons at risk for the outcome.

Secondary Outcome: Any spleen related intervention from time of study enrollment up to discharge.

Specific Aim #2: Determine factors related to failure of non-operatively managed blunt splenic injuries with splenic pseudoaneurysm or contrast blush or both.

Main Outcome: Splenectomy at anytime after study enrollment up to hospital discharge.

Analysis Method: Variables that may influence the in-hospital risk of splenectomy will be tested using univariate and multivariable regression using hierarchical modeling techniques. Because individual patients are clustered within hospitals, we will use hierarchical regression analysis to account for patient clustering. Hierarchical regression analysis is a statistical methodology that provides information on variables that are distributed between the individual and the hospital levels, quantifies the clustering of individual factors within hospitals, and permits the examination of cross level interactions between the effects of hospitals and individual level factors. Models will be built utilizing standard model building strategies. Factors that are associated with delayed splenectomy on univariate analysis (p<0.20) will be eligible for entry into the multivariable model. Factors with a p<0.05 on multivariable analysis will be considered independent predictors of delayed splenectomy after non-operative management of BSI.