Variability in Pediatric Splenic Injury Care: Results of a National Survey of General Surgeons

Background  Although nonoperative management is the standard of care for hemodynamically stable children with blunt splenic trauma, significant variation in practice exists. Little attention has been given to physician factors associated with management differences.

Design  Nationally representative mail survey conducted in June 2008.

Setting  United States.

Participants  Ten percent random sample of active, dues-paying fellows in the American College of Surgeons.

Main Outcome Measures  Knowledge, attitudes, and beliefs toward pediatric splenic injury management, including the role of clinical practice guidelines.

Results  Almost all of the 375 responding surgeons (97.4%) agreed that surgical intervention is not immediately necessary for hemodynamically stable children. However, surgeons reported significant disagreement regarding whether blood should be administered before operative intervention for hemodynamically unstable children and whether explorative surgery is needed for stable patients with evidence of contrast extravasation on computed tomography. Only 18.7% of surgeons reported being very familiar with the clinical practice guidelines for the management of pediatric blunt splenic trauma from either the Eastern Association for the Surgery of Trauma or the American Pediatric Surgical Association. Surgeons who were very familiar with either guideline were significantly more likely to rate the guidelines as beneficial (90.0% vs 72.8%, P = .002).

Conclusions  General surgeons reported varying degrees of familiarity with and use of clinical practice guidelines for pediatric splenic injury management. Limited pediatric experience and lack of pediatric hospital resources may limit more widespread adoption of nonoperative management. Targeted educational interventions may help increase surgeon knowledge of guidelines and best practices.

Management of children with hemodynamically stable traumatic spleen injury continues to shift from a predominantly operative to a predominantly nonoperative strategy. Current clinical practice guidelines recommend nonoperative management for hemodynamically stable children with blunt splenic trauma.^1-6 The Eastern Association for the Surgery of Trauma (EAST) recommends nonoperative management of blunt pediatric splenic injuries in hemodynamically stable patients irrespective of the splenic injury grade, citing associated lower overall morbidity and mortality.¹ The EAST guidelines also do not recommend routine imaging (computed tomography or ultrasound) for clinically improving, hemodynamically stable pediatric patients but suggest that angiographic embolization may be appropriate in the hemodynamically stable patient who continues to bleed. Similarly, the evidence-based guidelines of the American Pediatric Surgery Association (APSA) for children with isolated spleen injury do not recommend predischarge or postdischarge imaging for children with isolated spleen injuries and recommend intensive care unit stays only for children with grade IV splenic injuries.⁶

However, in some hospitals, performance of splenectomy is still common, with one study⁶ reporting as many as 61% of children receiving splenectomy. A multistate study⁷ of trauma and nontrauma centers reported splenectomy rates (15.3% and 19.3%, respectively) in excess of the APSA benchmarks for pediatric spleen injury management (5%-11% for children with spleen injuries and 0%-3% for children with isolated spleen injuries). Nonoperative management has also been shown to be more common in children's hospitals compared with general hospitals.⁸

The short-term benefits of nonoperative management of splenic injuries in hemodynamically stable children may include the avoidance of surgical costs, fewer blood transfusions, and shorter hospital lengths of stay.⁴ Spleen conservation also may help avoid infectious complications due to immunocompromise, including overwhelming postsplenectomy infection.^1,9,10 Spleen-conserving management may also produce long-term cost savings due to decreased prescription drug costs associated with prophylactic vaccinations and therapeutic antibiotics to reduce the likelihood of overwhelming postsplenectomy infection, as well as decreased hospitalization costs associated with this infection.

Numerous studies^7,8,11-13 have reported wide variability in pediatric spleen management, with differences observed by hospital type (children's hospital vs general hospital, for-profit hospital vs not-for-profit hospital, and trauma center vs nontrauma center). Conversely, physician factors associated with spleen management have received minimal attention.¹⁴ Factors, for example, the training and experience of the physician provider and the availability of pediatric surgeons and pediatric intensivists in the hospital, may affect the clinical decision to perform early splenectomy or to manage nonoperatively. Although a recent statewide quality improvement initiative in Washington State demonstrated an improvement in nonoperative management rates at pediatric and general trauma centers, general hospitals remain more likely to perform splenectomies in children with spleen injuries.¹⁵

In this study, we surveyed a nationally representative sample of general surgeons to identify common practices, attitudes, and knowledge regarding the treatment of blunt splenic injury in children. We assessed surgeon knowledge and the use of clinical practice guidelines and the perceived benefit of these guidelines in the care of these children.

We developed a questionnaire to address issues relevant to pediatric spleen management, such as (1) agreement or disagreement with approaches to care and treatment, (2) diagnostic tools used to evaluate these patients, (3) familiarity and use of guidelines for the management of blunt injury to the spleen from the EAST and the APSA, (4) specific management issues, (5) barriers limiting the use of nonoperative management, and (6) demographics.

With permission from the American College of Surgeons, we accessed the American College of Surgeons mailing database to generate a 10% random sample of current, active, dues-paying general surgeons who practice in the United States. We excluded surgical subspecialists, including pediatric surgeons. A sample of 1379 general surgeons was created for this study, and a cover letter and questionnaire were mailed to each surgeon in June 2008. A second contact consisted of a thank you/reminder card that was sent 1 week after the initial mailing. The final contact consisted of a cover letter and replacement questionnaire sent to all nonrespondents approximately 4 weeks after the initial mailing. The study protocol was reviewed by the institutional review board of the University of Arkansas for Medical Sciences and was deemed exempt.

Unidentifiable data were entered into a Microsoft Access (Microsoft Corp, Redmond, Washington) database. Univariate and bivariate analyses were completed using the t tests for continuous variables and the χ² tests for categorical variables. To identify significant associations between physician factors and knowledge of, or use of, clinical practice guidelines, we used logistic regression. All analyses were conducted using STATA/MP 10.0 (StataCorp LP, College Station, Texas). Explanatory variables are presented as odds ratios (ORs) with confidence intervals (CIs) and P values.

Of the 1379 surgeons contacted, 578 (41.9% unadjusted) returned surveys. Of the returned surveys, 166 were ineligible and, thus, were not completed for reasons, such as no longer practicing, deceased, or having a practice limited to adult patients. In addition, 28 respondents returned blank surveys with explicit refusal, and 9 questionnaires were returned with unknown eligibility due to undeliverable addresses or other. In the final analysis, we included data from 375 completed questionnaires, with an adjusted response rate of 32%. Table 1 provides demographic information about the sample.

We asked surgeons whether they agree or disagree with 4 statements pertaining to the management of pediatric splenic injury (Table 2). Overall, more than 97% of surgeons agreed with the statement that surgical treatment is not immediately necessary for hemodynamically stable children with blunt splenic injuries. However, surgeons reported significant disagreement regarding whether blood should be administered before operative intervention for hemodynamically unstable children, with 24.5% disagreeing with this statement and another 15% neither agreeing nor disagreeing. Whether surgeons believe that explorative surgery is needed for stable patients with evidence of contrast extravasation on computed tomography was also more varied, with 34.8% either agreeing or reporting neutral answers. Similarly, considerable variation was reported on whether hemodynamically stable children require an admission to the intensive care unit.

In the Figure, we present the proportion of general surgeons who reported to be very familiar with EAST or APSA guidelines for the management of pediatric splenic trauma. In all surgeon age groups, less than 25% of general surgeons reported that they are very familiar with EAST or APSA guidelines. Familiarity with EAST guidelines was greater than that for APSA guidelines in all age groups, with less than 10% of general surgeons reporting that they are very familiar with EAST or APSA guidelines. Compared with urban practices, general surgeons who practice in rural settings were less likely to report being very familiar with the guidelines (8.5% vs 26.4%, P = .001). Similarly, community general surgeons were less likely to report being very familiar than were general surgeons practicing in academic settings (13.1% vs 52.8%, P < .001). No difference was observed by surgeon sex.

In the multivariable logistic regression, we identified independent predictors of familiarity with the guidelines, controlling for potential confounders (Table 3). Age and sex were not significant predictors of familiarity with the guidelines. Location of practice setting and type of hospital were associated with guideline familiarity. Compared with general surgeons practicing in academic hospitals, general surgeons in large town community hospitals and small rural community hospitals were much more likely to report not being very familiar with either EAST or APSA guidelines (OR, 3.01; 95% CI, 1.23-7.38; P = .02 and 8.20; 3.07-21.89; P < .001, respectively). General surgeons practicing in small rural community hospitals were significantly more likely to report not often using the guidelines in their practice (OR, 3.48; 95% CI, 1.67-7.29; P = .001).

We asked general surgeons to identify which diagnostic tools are typically used in their practice setting to evaluate the abdomen of a pediatric patient after blunt trauma. Table 4 provides the results of this question, stratified by guideline familiarity. For hemodynamically stable patients, approximately 97% of surgeons reported diagnostic use of computed tomography, with no difference by guideline familiarity or use. Similarly, the use of diagnostic peritoneal lavage is consistently uncommon for hemodynamically stable children with blunt splenic injury regardless of guideline familiarity or use. We observed less use of radiography and more use of focused abdominal sonography for trauma by surgeons who reported that they often use the guidelines. For hemodynamically unstable patients, surgeons who are very familiar with and often use the guidelines were much more likely to report using focused abdominal sonography for trauma for diagnostic purposes.

To better understand the issues that affect operative vs nonoperative management decisions, we asked surgeons to identify barriers and factors that limit the use of nonoperative management of pediatric splenic trauma in their hospitals (Table 5). The most commonly identified major barrier or factor was not having a pediatric intensive care unit (30.9%), followed by limited pediatric experience by the surgeon (18.7%). Reimbursement (physician or hospital) issues do not seem to be a significant limiting factor. Approximately 1 in 5 surgeons reported practice convenience and lack of surgical coverage as factors (major or minor) that limit nonoperative management.

We asked surgeons to report how beneficial guidelines are for the management of blunt splenic injury in pediatric patients. Overall, 81% of surgeons reported that the guidelines are either somewhat or very beneficial. Compared with surgeons who were not very familiar with the guidelines, general surgeons reporting high familiarity were more likely to report a beneficial value of the guidelines (90.0% vs 72.8%, P = .002). Of surgeons who reported using the guidelines often, 96.3% reported that the guidelines are beneficial.

Nonoperative or spleen-conserving management of hemodynamically stable children with splenic injuries is the standard of care. In this study of a broad, representative sample of US general surgeons, we identified variability in the practices and beliefs pertaining to the management of blunt splenic trauma in children. We discovered differences in agreement in the management of these children, including variability in the usual diagnostic approaches to evaluating these cases. To our knowledge, this is the first study to examine general surgeon knowledge and practices in this area.

Rural surgeons are less likely to be familiar with the guidelines and are less likely to use them in practice. This may reflect, in part, the difficulties and challenges of practicing in rural communities. In addition, the low volume of pediatric trauma cases seen by surgeons in small rural hospitals might lead to less priority being placed on continuing education in this area of management. Given the low frequency of such cases in rural settings but the high degree of variability in management and the serious outcomes that may result from deviation from best practice, one might argue that these guidelines should be more specifically emphasized or made available to rural surgeons. However, the mechanisms and feasibility for providing such continuing education in rural settings are limited owing to high clinical workloads, geographic isolation, and other factors. Many small rural hospitals are designated under the federal Critical Access Hospital program, allowing cost-based reimbursement to the hospitals but not to their physician providers. Although cost-based reimbursement may help keep these hospitals open, it is unclear how the rural surgeon fares under this program. One critical question is whether appropriate incentives, economic or otherwise, are in place to encourage continuing education on topics that are infrequently encountered in rural surgical practice, including pediatric spleen management.

Studies have documented lower splenectomy rates in children cared for (1) at pediatric trauma centers in Pennsylvania,¹² (2) by pediatric surgeons than by general surgeons,¹⁶ and (3) in pediatric hospitals than in general hospitals.⁸ These studies typically focused on urban settings because few rural communities can provide the resources to support access to pediatric surgeons, the availability of pediatric intensive care units, or the development and qualification of pediatric trauma centers. Given this dilemma, one reasonable strategy is to enhance clinical collaboration between rural general surgeons and pediatric trauma centers or pediatric hospitals using resources, such as teleradiology for timely and cost-effective real-time consultation, logistically feasible treatment algorithms based on EAST or APSA guidelines, and established transfer agreements, all of which would potentially contribute to an increased frequency of spleen-conserving management.

The 2 areas of greatest variability in responses involved the willingness to consider transfusion before operative intervention and the importance of active contrast extravasation on computed tomography in decision making regarding operative intervention. These are areas in which additional development of existing guidelines may help standardize this approach. Recent studies³ have suggested that it is reasonable to consider transfusion of up to 10 mL/kg before considering splenectomy, and, as a result, many pediatric centers are moving in that direction. Contrast extravasation on computed tomography has been associated with higher rates of failure of nonoperative management in adults, but recent studies^17,18 have also demonstrated that significant splenic salvage rates can be achieved in this patient group, especially in children.

There are several limitations to the present study. First, we cannot rule out the potential for nonresponse bias in the survey. Differences may exist between surgeons who responded to the survey and those who did not. Second, the clinical decision to perform splenectomy vs attempts at conservative management of splenic injury may depend on the training and experience of the physician and on the availability of pediatric hospital resources, including pediatric specialists. Third, we could not assess pediatric surgery experience among the respondents, and assessing the pediatric capabilities of the hospitals was beyond the scope of this study.

In summary, this study offers new information about the physician factors associated with the management of pediatric blunt splenic trauma. Evidence-based clinical practice guidelines are becoming ever more prevalent. Focused strategies to ensure the adoption of spleen-conserving management seem to be needed, such as disseminating the pediatric spleen management guidelines to community surgeons and hospital trauma programs, emphasizing pediatric spleen management during trauma designation reviews by state oversight agencies and during verification visits by the American College of Surgeons, disseminating hospital-specific pediatric spleen management performance data to highlight areas with potential for improvement, and ongoing review of pediatric spleen management data to ensure compliance with the guidelines.

Correspondence: Stephen M. Bowman, PhD, Department of Health Policy and Management, Johns Hopkins University, 624 N Broadway, Room 539, Baltimore, MD 21205-1996 (smbowman@jhsph.edu).

Accepted for Publication: September 2, 2009.

Author Contributions: Dr Bowman had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Bowman, Bulger, Sharar, and Smith. Acquisition of data: Bowman and Maham. Analysis and interpretation of data: Bowman, Bulger, Sharar, and Smith. Drafting of the manuscript: Bowman, Sharar, Maham, and Smith. Critical revision of the manuscript for important intellectual content: Bowman, Bulger, Sharar, and Smith. Statistical analysis: Bowman. Obtained funding: Bowman. Administrative, technical, and material support: Bowman and Maham. Study supervision: Bowman and Bulger.

Financial Disclosure: None reported.

Funding/Support: This research was supported, in part, by the Arkansas Children's Hospital Research Institute (Dr Bowman) and by the Arkansas Biosciences Institute, the major research component of the Arkansas Tobacco Settlement Proceeds Act of 2000 (Dr Bowman).