Ensuring scientific reproducibility in bio-macromolecular modeling via extensive, automated benchmarks

Each year vast international resources are wasted on irreproducible research. The scientific community has been slow to adopt standard software engineering practices, despite the increases in high-dimensional data, complexities of workflows, and computational environments. Here we show how scientific software applications can be created in a reproducible manner when simple design goals for reproducibility are met. We describe the implementation of a test server framework and 40 scientific benchmarks, covering numerous applications in Rosetta bio-macromolecular modeling. High performance computing cluster integration allows these benchmarks to run continuously and automatically. Detailed protocol captures are useful for developers and users of Rosetta and other macromolecular modeling tools. The framework and design concepts presented here are valuable for developers and users of any type of scientific software and for the scientific community to create reproducible methods. Specific examples highlight the utility of this framework, and the comprehensive documentation illustrates the ease of adding new tests in a matter of hours.

Development and Evaluation of GlycanDock: A Protein-Glycoligand Docking Refinement Algorithm in Rosetta

Carbohydrate chains are ubiquitous in the complex molecular processes of life. These highly diverse chains are recognized by a variety of protein receptors, enabling glycans to regulate many biological functions. High-resolution structures of protein-glycoligand complexes reveal the atomic details necessary to understand this level of molecular recognition and inform application-focused scientific and engineering pursuits. When experimental challenges hinder high-throughput determination of quality structures, computational tools can, in principle, fill the gap. In this work, we introduce GlycanDock, a residue-centric protein-glycoligand docking refinement algorithm developed within the Rosetta macromolecular modeling and design software suite. We performed a benchmark docking assessment using a set of 109 experimentally determined protein-glycoligand complexes as well as 62 unbound protein structures. The GlycanDock algorithm can sample and discriminate among protein-glycoligand models of native-like structural accuracy with statistical reliability from starting structures of up to 7 Å root-mean-square deviation in the glycoligand ring atoms. We show that GlycanDock-refined models qualitatively replicated the known binding specificity of a bacterial carbohydrate-binding module. Finally, we present a protein-glycoligand docking pipeline for generating putative protein-glycoligand complexes when only the glycoligand sequence and unbound protein structure are known. In combination with other carbohydrate modeling tools, the GlycanDock docking refinement algorithm will accelerate research in the glycosciences.

PyRosetta Jupyter Notebooks Teach Biomolecular Structure Prediction and Design

Biomolecular structure drives function, and computational capabilities have progressed such that the prediction and computational design of biomolecular structures is increasingly feasible. Because computational biophysics attracts students from many different backgrounds and with different levels of resources, teaching the subject can be challenging. One strategy to teach diverse learners is with interactive multimedia material that promotes self-paced, active learning. We have created a hands-on education strategy with a set of sixteen modules that teach topics in biomolecular structure and design, from fundamentals of conformational sampling and energy evaluation to applications like protein docking, antibody design, and RNA structure prediction. Our modules are based on PyRosetta, a Python library that encapsulates all computational modules and methods in the Rosetta software package. The workshop-style modules are implemented as Jupyter Notebooks that can be executed in the Google Colaboratory, allowing learners access with just a web browser. The digital format of Jupyter Notebooks allows us to embed images, molecular visualization movies, and interactive coding exercises. This multimodal approach may better reach students from different disciplines and experience levels as well as attract more researchers from smaller labs and cognate backgrounds to leverage PyRosetta in their science and engineering research. All materials are freely available at https://github.com/RosettaCommons/PyRosetta.notebooks.

Novel sampling strategies and a coarse-grained score function for docking homomers, flexible heteromers, and oligosaccharides using Rosetta in CAPRI rounds 37-45

Critical Assessment of PRediction of Interactions (CAPRI) rounds 37 through 45 introduced larger complexes, new macromolecules, and multistage assemblies. For these rounds, we used and expanded docking methods in Rosetta to model 23 target complexes. We successfully predicted 14 target complexes and recognized and refined near-native models generated by other groups for two further targets. Notably, for targets T110 and T136, we achieved the closest prediction of any CAPRI participant. We created several innovative approaches during these rounds. Since round 39 (target 122), we have used the new RosettaDock 4.0, which has a revamped coarse-grained energy function and the ability to perform conformer selection during docking with hundreds of pregenerated protein backbones. Ten of the complexes had some degree of symmetry in their interactions, so we tested Rosetta SymDock, realized its shortcomings, and developed the next-generation symmetric docking protocol, SymDock2, which includes docking of multiple backbones and induced-fit refinement. Since the last CAPRI assessment, we also developed methods for modeling and designing carbohydrates in Rosetta, and we used them to successfully model oligosaccharide-protein complexes in round 41. Although the results were broadly encouraging, they also highlighted the pressing need to invest in (a) flexible docking algorithms with the ability to model loop and linker motions and in (b) new sampling and scoring methods for oligosaccharide-protein interactions.

Prospective randomized observer-blinded study comparing the analgesic efficacy of ultrasound-guided rectus sheath block and local anaesthetic infiltration for umbilical hernia repair

BACKGROUND

Umbilical hernia repair, a common day-surgery procedure in children, is associated with considerable postoperative discomfort. Possible modes of postoperative analgesia for umbilical hernia repair are rectus sheath block (RSB) and local anaesthetic infiltration of the surgical site (LAI).

METHODS

We undertook an observer-blinded, randomized, prospective, observational study to compare the efficacy of ultrasound-guided RSB and LAI in providing postoperative analgesia for umbilical hernia repair. Our primary objective was to compare the use of opioid medication between patients who receive RSB and those who receive LAI. Our secondary objectives were to compare the duration of analgesia based on time to first rescue analgesic, to compare the quality of analgesia based on revised FACES scale, and to determine the incidence of side-effects.

RESULTS

Fifty-two patients (26 in each group) completed the study. There was a statistically significant difference in the perioperative opioid medication consumption between the LAI group [mean: 0.13 mg kg(-1), confidence interval (0.09-0.17 mg kg⁻¹)] and the RSB group [mean: 0.07 mg kg⁻¹, confidence interval (0.05-0.09 mg kg⁻¹)] (P=0.008). When we compared the postoperative opioid consumption between the LAI group [mean: 0.1 mg kg⁻¹, 95% confidence interval (0.07-0.13 mg kg⁻¹)] and the RSB group [mean: 0.07 mg kg(-1), 95% confidence interval (0.05-0.09 mg kg⁻¹)] (P=0.09), there was a trend towards statistical significance between the two groups. The difference in time to rescue analgesic administration between the RSB group [49.7 (36.9) min] and the LAI group [32.4 (29.4) min] was not statistically significant (P=0.11).

CONCLUSIONS

This study demonstrates that ultrasound-guided RSB provides superior analgesia in the perioperative period compared with infiltration of the surgical site after umbilical hernia repair. In comparing only the postoperative period, analgesia provided by an ultrasound-guided RSB showed a trend towards statistically significant improvement compared with infiltration of the surgical site.

Determining objective injury prevention priorities

BACKGROUND AND OBJECTIVE

Finite injury prevention resources make the establishment of prevention priorities essential. Toward this end, the US National Trauma Data Bank (NTDB) for 2000 to 2004 was accessed and four injury prevention priority scores (one previously defined and three new scores) were computed.

METHODS

An injury prevention priority score (IPPS) was calculated based on the frequency of an injury mechanism and the median injury severity score. In addition, a mortality priority score (Mort-PS), a hospital charge priority score (Charge-PS), and a years of potential life lost (YPLL-PS) priority score were calculated for the 13 most common injury mechanisms.

RESULTS

There was variability across the four scores, but motor vehicle traffic, firearm related, and fall injuries ranked high on all four of the priority criteria. Multiple criteria should be considered when assessing injury burden.

CONCLUSIONS

The methods presented here can help prioritize injuries and support more objective public policies.

Management of duodenal injuries in children

BACKGROUND/PURPOSE

The natural history and management of pediatric duodenal injuries are incompletely described. This study sought to review injury mechanism, surgical management, and outcomes from a collected series of pediatric duodenal injuries.

METHODS

A retrospective chart review was conducted for a 10-year period of all children less than 18 years old treated for duodenal injuries at 2 pediatric trauma centers.

RESULTS

Forty-two children were treated for duodenal injuries. There were 33 blunt and 9 penetrating injuries. Injuries were classified using the Organ Injury Scale for the Duodenum. Twenty-four patients underwent operative management by primary repair (18), duodenal resection and gastrojejunostomy (4), or pyloric exclusion (2). Duodenal hematomas were treated nonoperatively in 94% of cases. The average ISS for operative versus nonoperative cases was 23 and 10, respectively. Delay in diagnosis or operative intervention (>24 hours) was associated with increased complication rate (43% v 29%) and hospitalization (32 v 20 days). Nine children requiring surgery experienced delays and were most highly associated with foreign body, child abuse, and bicycle injuries. There were no deaths caused by duodenal injuries.

CONCLUSIONS

Duodenal injuries in children were predominantly blunt and had a low mortality rate. When surgery was required, primary repair was usually feasible.

Nonoperative management of solid abdominal organ injuries from blunt trauma: impact of neurologic impairment

The role of nonoperative management of solid abdominal organ injury from blunt trauma in neurologically impaired patients has been questioned. A statewide trauma registry was reviewed from January 1993 through December 1995 for all adult (age >12 years) patients with blunt trauma and an abdominal solid organ injury (kidney, liver, or spleen) of Abbreviated Injury Scale score > or =2. Patients with initial hypotension (systolic blood pressure <90 mm Hg) were excluded. Patients were stratified by Glasgow Coma Score (GCS) into normal (GCS 15), mild to moderate (GCS 8-14), and severe (GCS < or =7) impairment groups. Management was either operative or nonoperative; failure of nonoperative management was defined as requiring laparotomy for intraabdominal injury more than 24 hours after admission. In the 3-year period 2327 patients sustained solid viscus injuries; 1561 of these patients were managed nonoperatively (66 per cent). The nonoperative approach was initiated less frequently in those patients with greater impairment in mental status: GCS 15, 71 per cent; GCS 8 to 14, 62 per cent; and GCS < or =7, 50 per cent. Mortality, hospital length of stay, and intensive care unit days were greater in operatively managed GCS 15 and 8 to 14 groups but were not different on the basis of management in the GCS < or =7 group. Failure of nonoperative management occurred in 94 patients (6%). There was no difference in the nonoperative failure rate between patients with normal mental status and those with mild to moderate or severe head injuries. Nonoperative management of neurologically impaired hemodynamically stable patients with blunt injuries of liver, spleen, or kidney is commonly practiced and is successful in more than 90 per cent of cases. No differences were noted in the rates of delayed laparotomy or survival between normal, mild to moderately head-injured, and severely head-injured patients.

Severe blunt hepatic trauma in children

BACKGROUND

Severe blunt hepatic injury in children is associated with a high mortality rate. Although nonoperative management has become the treatment of choice for mild to moderate liver trauma, there is no consensus as to the optimal treatment for the most severe hepatic injuries in children.

METHODS

A statewide trauma registry was reviewed to identify children (age 18 years or less) treated for a severe blunt liver injury for the period 1993 to 1998. Only children with an American Association for the Surgery of Trauma grade V (AIS code 541828.5) liver injury were included. Database records were reviewed for demographic information, associated injuries, survival rate, length of stay (LOS), intensive care days (ICUD), and treatment rendered after resuscitation in the emergency department.

RESULTS

Thirty children with a grade V liver injury were identified. The mean age was 11.2 years (range, 1 to 18), and the overall survival rate was 56%. Data for 5 patients were excluded (4 patients died in the emergency department, and 1 patient was transferred to another institution after arrival). Survivors had a trend toward a lower injury severity score (ISS) (36.1 v 44.6; P <.1) and a significantly higher Glasgow Coma Scale (GCS), 12.5 v 6.6; P <.007). Patients with a decreased GCS had a lower overall survival rate (GCS < 8, 30% v GCS > 8, 76%). In the subset of 14 patients taken directly to the operating room, there was no difference between survivors (n = 6, 43%) and nonsurvivors (n = 8, 57%) in ISS (43 v 43; P value, not significant) or GCS (8.6 v 8.0; P value, not significant). Of the 11 patients treated nonoperatively, 10 (91%) survived with an average ISS of 33 and GCS of 13.8. Nonsurvivors more often had identified associated injuries to other abdominal and retroperitoneal organs.

CONCLUSIONS

Severe hepatic injury is associated with a very high overall mortality rate in children. A low GCS is associated with a significant decrease in survival rate and may be the most important factor in outcome. Patients taken directly to the operating room have a slightly greater injury severity and a decreased survival rate compared with those treated nonoperatively. Thresholds and indications for laparotomy in these patients are not clear, and the need for operative management should be guided by the child's physiologic response to resuscitation. For those patients whose physiologic response to resuscitation permitted nonoperative management, a good outcome was achieved.

The New Injury Severity Score and the evaluation of pediatric trauma

BACKGROUND

To compare the effectiveness of the Injury Severity Score (ISS) and New Injury Severity Score (NISS) in predicting mortality in pediatric trauma patients.

METHODS

NISS, the sum of the squares of a patient's three highest Abbreviated Injury Scale scores (regardless of body region), were calculated for 9,151 patients treated at four regional pediatric trauma centers and compared with previously calculated ISS values. The power of the two scoring systems to predict mortality was gauged through comparison of misclassification rates, receiver operating characteristic curves, and Hosmer-Lemeshow goodness-of-fit statistics.

RESULTS

Although there were significant differences in mean NISS and ISS values for each hospital, differences in the predictive abilities of the two scoring systems were insignificant, even when analysis was restricted to the subgroup of patients with severe or penetrating injuries.

CONCLUSION

The significant differences in the predictive abilities of the ISS and NISS reported in studies of adult trauma patients were not seen in this review of pediatric trauma patients.

Predicting hollow visceral injury in the pediatric blunt trauma patient with solid visceral injury

BACKGROUND

Nonoperative management of a solid organ injury (SVI) is accepted in the stable pediatric trauma patient. A concern with nonoperative management is missing a hollow visceral injury (HVI). Factors that may help predict HVI have not been well documented.

METHODS

The National Pediatric Trauma Registry was reviewed for the period October 1988 through September 1998 for all blunt injured, hemodynamically stable pediatric patients (age < or =12 years) with an SVI (kidney, liver, pancreas, spleen) of Abbreviated Injury Scale (AIS) score > or =2. HVIs included AIS > or =2 gastrointestinal tract injuries.

RESULTS

For the decade of review, 2,977 pediatric patients sustained an SVI, including 96 with an HVI (3.2%). The mean age was 6.6 years, with a mean Injury Severity Score of 12.4. An occupant in a motor vehicle accident was the most common injury mechanism (30.4%), but assault was the most likely to result in an HVI (11.5%). The liver was the most common SVI (n = 1,400), the spleen the least likely to have an associated HVI (2.5%). Pancreatic injuries had a higher rate of HVI (P < .001). The majority of patients had a single SVI (n = 2,507) with 71 associated HVIs (2.8%). The risk of associated HVI increased as the number of solid organs injured increased: 4.7% with 2 organs, 13.5% if 3 organs were injured (P< .001). In patients with a single SVI, the rate of HVI did not differ as AIS increased (range, 2.7% to 6.5%, Pvalue not significant).

CONCLUSIONS

The overall rate of HVI was low (3.2%). Higher rates of HVI were found in assaulted patients and patients with multiple SVIs or pancreatic injuries. The risk of associated HVI was dependent more on number of SVIs than severity of the individual organ injury. This data suggest that nonoperative management is justified in the patient with a single SVI but should be used cautiously in the patient with multiple SVI or a pancreatic injury.

Appendicitis in the young child: a continuing diagnostic challenge

OBJECTIVE

The purpose of this review was to examine the presenting signs and symptoms of children 5 years of age or less who underwent operation for appendicitis. In addition, we sought to determine the rate of perforation of the appendix and the effect on outcome in this age group.

METHODS

Medical records for the period September 1987 to September 1998 were reviewed for all children 5 years of age or less who underwent appendectomy for appendicitis. Data gathered included age at operation, gender, care sought prior to admission for appendectomy, duration of symptoms, signs and symptoms at the time of admission, and length of postoperative hospital stay. Symptoms of diarrhea, emesis, fever, pain, and anorexia were recorded. Physical signs of an abdominal mass, guarding, rebound tenderness, rigidity, and diffuse or focal tenderness were recorded. Diagnostic information included white blood cell count with differential, and radiographic imaging, if obtained. The presence or absence of perforation of the appendix, and abscess formation were based on the intraoperative impression of the operating surgeon.

RESULTS

For the 11-year period, 120 patients 5 years of age or less required an operation for appendicitis and had a complete medical database. The mean age was 3.6 +/- 1.3 years; 53% were male. Patients underwent a separate medical evaluation prior to arriving at a definitive diagnosis in 44.2 % cases. The most common presenting symptom was abdominal pain (94%); the most common sign was abdominal tenderness (95.8%). Tenderness was generally diffuse if perforation had occurred (62%) or focal in the nonperforated group (61%). The duration of symptoms in patients with perforation was more than double that of the nonperforated patients (4.7 vs 2.1 days, respectively). The mean white blood cell count (WBC) was 18.3 +/- 7.4 cells/mm3, and did not differ significantly between the perforated and nonperforated groups. A left shift detected in the WBC differential was present in 91%. An abdominal radiograph was obtained in 87%, and demonstrated a fecalith in 18%. A preoperative ultrasound was obtained in 38%, a computed tomographic scan in 7%. At the time of surgery, 74% were found to have evidence of perforation. An abscess was found at the initial surgery in 47% of patients with appendiceal perforation, but in no patient in whom perforation had not occurred. The rate of perforation increased as the age of the patient decreased (100% perforation for age 1 (n = 10) to 69% for age 5, (n = 35). Perforation was associated with a longer hospital length of stay as compared to the nonperforated appendix (median 9 days vs. 3 days, respectively, P < 0.001). There were no deaths in this series.

CONCLUSION

Appendiceal perforation continues to be a common occurrence in the young child and increases in frequency as the age of the patient decreases and the duration of symptoms lengthens. Perforation results in a significant increase in hospital length of stay and rate of abscess formation.

Congenital hyperinsulinism and the surgeon: lessons learned over 35 years

BACKGROUND/PURPOSE

Congenital hyperinsulinism induces severe and unremitting hypoglycemia in newborns and infants. If poorly controlled, seizures and irreversible brain damage may result. Subtotal (<95%) or near-total (95% to 98%) pancreatectomy have been performed for glycemic control in babies who do not respond to aggressive medical therapy. Because hypoglycemia often persists after subtotal resection, 95% pancreatectomy has emerged as the procedure of choice. To define the effect of more or less extensive pancreatectomy on the management and outcome of refractory congenital hyperinsulinism, the authors examined our single institutional experience.

METHODS

The records of children treated between 1963 and 1998 for congenital hyperinsulinism, and who required pancreatectomy, were reviewed. Outcome parameters included glycemic response to surgery, need for reresection, surgical morbidity, surgical and long-term mortality, and development of diabetes mellitus (DM). A complete response was defined as discharge to home on no glycemic medications, no continuous feedings, and without DM. Histological reports were reviewed and categorized as either diffuse or focal disease.

RESULTS

Of 101 children treated for congenital hyperinsulinism during this period, 53 (50%) required pancreatectomy for glucose control. Mean follow-up for the study population was 9.8 +/- 1.1 years. Overall, 23 children (43%) showed a complete response, occurring in 50% of patients having > or = 95% pancreatectomy (n = 34), but in only 19% having less than 95% resection (n = 16). The remaining three babies had local excision of a solitary focal lesion, and each showed a complete response. Histopathology showed diffuse islet abnormalities in 42 specimens (79%) and solitary focal lesions in 11 (21%). A complete response was observed for 82% of focal but only 33% of diffuse lesions. Eight patients (15%) required reresection for persistent hypoglycemia, seven having diffuse lesions and one focal. Surgical morbidity occurred in 13 cases (26%), and the 30-day surgical mortality rate was 6%, each death (n = 3) occurring before 1975. DM developed in seven children (14%), each having diffuse lesions, and was independent of resection type.

CONCLUSION

Because euglycemia is more readily restored, and because the risks for surgical complications and DM do not appear increased, the authors recommend 95% pancreatectomy as the initial procedure of choice for newborns and infants with congenital hyperinsulinism.

Thickening of the diaphragm: a new computed tomography sign of diaphragm injury

Computed tomography examination is becoming a standard method of evaluating injury in blunt trauma. This report describes a previously unreported computed tomography sign of diaphragmatic injury. A retrospective review of imaging findings was performed on eight patients with surgically proven traumatic diaphragmatic rupture. Thickening of the diaphragm was present on computed tomography in six of eight patients with surgically proven diaphragm rupture. Two patients demonstrated previously established computed tomography signs of diaphragm rupture and two patients had no findings to suggest diaphragm injury. Diaphragm thickening is associated with diaphragm rupture and may be a useful diagnostic sign in the evaluation of patients with blunt trauma.

Solid viscus injury predicts major hollow viscus injury in blunt abdominal trauma

BACKGROUND

As nonoperative management of blunt abdominal trauma has become more popular, reliable models for predicting the likelihood of concomitant hollow viscus injury in the hemodynamically stable patient with a solid viscus injury are increasingly important.

METHODS

The Pennsylvania Trauma Systems Foundation registry was reviewed for the period from January 1992 to December 1995 for all adult (age > 12 years) patients with blunt trauma and an Abbreviated Injury Scale (AIS) score > or = 2 for a solid viscus (kidney, liver, pancreas, spleen). Patients with an initial systolic blood pressure < 90 mm Hg were excluded. Hollow viscus injuries included only lacerations or perforations of the gallbladder, gastrointestinal tract, or urinary tract.

RESULTS

In the 4-year period, 3,089 patients sustained solid viscus injuries, 296 of whom had a hollow viscus injury (9.6%). The mean age was 35.6 years, mean Injury Severity Score was 22.2, and mean Revised Trauma Score was 7.3; 63.3% of the patients were male. A solitary solid viscus injury occurred in 2,437 patients (79%), 177 of whom (7.3%) had a hollow viscus injury. The frequency of hollow viscus injury increased with the number of solid organs injured: 15.4% of patients with two solid viscus injuries (n = 547) and 34.4% of patients with three solid viscus injuries (n = 96) suffered a concomitant hollow viscus injury (p < 0.001 vs. one organ). A hollow viscus injury was 2.3 times more likely for two solid viscus injuries and 6.7 times more likely for three solid viscus injuries compared with a solitary solid viscus injury. For solitary solid viscus injury, the frequency of hollow viscus injury varied little with increasing AIS score (AIS score 2, 6.6%; AIS score 3, 8.2%; AIS score 4, 9.2%; AIS score 5, 6.2%) (p = 0.27 between groups), suggesting that the incidence of hollow viscus injury is related more to the number of solid visceral injuries than the severity of individual organ injury. Also, when the sum of the AIS scores for solid viscus injuries was <6, the mean rate of hollow viscus injury was 7.8%. This increased to 22.8% when the sum of the AIS scores for solid viscus injury was > or =6 (p < 0.001). A pancreatic injury in combination with any other solid viscus injury had a rate of hollow viscus injury of >33%.

CONCLUSION

A model of organ injury scaling predicted hollow viscus injury. Multiple solid viscus injuries, particularly pancreatic, or abdominal solid viscus injuries with an AIS score > or = 6, were predictive of hollow viscus injury. Identification of these injury patterns should prompt consideration for early operative intervention.

Firearm injury among urban youth during the last decade: an escalation in violence

PURPOSE

To better characterize firearm violence in urban youth, the authors investigated the circumstances and outcome of shootings among youths under 17 years of age.

METHODS

A retrospective case review was undertaken of all patients under 17 years of age treated for a gunshot wound at two adjoining level I Trauma Centers (adult and pediatric) administering to a predominantly lower socioeconomic population from January 1986 to December 1995. Demographics, injury severity, circumstances, and outcome of injury were analyzed.

RESULTS

323 youths were wounded by firearms. The mean age was 12.8 years, and 82.3% were boys. There was a 110% increase in frequency of wounding noted during the second half of study (219 v 104), predominantly in the adolescent subset (160% increase for age greater than 12 years v 30% increase for age 12 years and under). The mean injury severity score and trauma score were 9.3 and 14.4, respectively. Violent circumstances (assault, crossfire, drive-by shooting, suicide) accounted for 60.4% of injuries and more than doubled over the study period (26.7% of total in the first 2 years [n = 4], 68.8% in the last 2 years [n = 55]). Unintentional injuries (self nonsuicide, family, friend) accounted for 26.3% of the injuries and declined in relative frequency over the study period (46.7% of total in the first 2 years [n = 7], 20.0% in the last 2 years [n = 16]). Black boys had the highest wounding incidence (9.2/1,000 population), were most commonly injured by assault (29.0%), and had a higher mean number of wounds (1.8). White boys had a lower wounding incidence (3.1/1,000 population), were more often injured unintentionally by a friend (41.2%), and had a lower mean number of wounds (1.3), none as a result of violence. Girls had a wounding incidence of less than 2/1,000 and were most commonly injured in crossfire (40.7%). Ten percent of shootings were fatal. The assailant was known to the victim in 52.8% of children less than 6 years of age, but only 24.7% of children over 12 years of age.

CONCLUSIONS

The incidence of gunshot wounds in the youth of this urban population has increased substantially over the past decade. Adolescent black boys were the most frequent victims of these shootings. There has been a disproportionate growth in violent circumstances surrounding the shootings.

Shotgun wounds in children. Not just accidents

OBJECTIVE

To characterize the demographic characteristics of shotgun wounds in children and adolescents across various regions within a state.

DESIGN

Retrospective case study.

SETTING

Accredited trauma centers in Pennsylvania.

PATIENTS

All patients less than 18 years old who sustained shotgun wounds.

DATA COLLECTION

Patient data were collected from the Pennsylvania Trauma Systems Foundation statewide trauma registry for January 1987 through December 1994. Data reviewed included age, race, sex, region, nature of injury, assailant, location of incident, length of stay in the hospital and intensive care unit, Injury Severity Score, organs injured, death, and discharge disposition.

RESULTS

Over 8 years there were 95 shotgun wounds in patients with a mean +/- SD age of 14.0 +/- 3.7 years; the male-to-female ratio was 5.8:1. The incidence of shotgun wounds in urban areas increased threefold during the second half of the study; the incidence in nonurban regions was unchanged. Eighteen deaths (19%) occurred, 17 (94%) within 24 hours and 10 (56%) because of intracranial injury. Overall, unintentional shotgun wounds were most common (n = 46 [48%]), followed by assaults (n = 37 [39%]) and suicides (n = 8 [8%]). The highest per capita incidence of shotgun wounds occurred in urban areas, typically the result of an assault (n = 30 [73%]). In nonurban areas, shotgun wounds were usually unintentional (n = 36 [67%]); 34 (63%) occurred in the home. In contrast, in urban areas, 26 shotgun wounds (63%) occurred on the street. Overall, 14 fatal shotgun wounds (78%) occurred in the home. Operative intervention was required for 57 patients (60%). Ultimately, 67 patients (71%) were discharged to home.

CONCLUSIONS

In urban areas, shotgun wounds are increasing in incidence, often occur on the street, and often result from assault. In nonurban areas, shotgun wounds are usually unintentional, often occur in the home, and are more often lethal than shotgun wounds in urban areas. Multiple-organ injury, surgery, and lengthy hospital stays are common.

Thoracic gunshot wounds in children under 17 years of age

Over the past decade there has been a significant increase in the incidence of gunshot wounds (GSW) among Americans under 19 years of age. Despite the increase, pediatric GSW have received little attention in the literature, and no study has focused on pediatric victims of thoracic GSW. In the present study, the authors performed a retrospective review of the records of 51 patients under 17 years of age who were treated for thoracic GSW at adjoining level I pediatric and adult trauma centers between July 1987 and June 1995. The primary catchment area for these institutions is a lower socioeconomic, urban tenancy. The study population was 80.4% male and 86.3% black; the mean age was 12.4 years. The mean injury severity score and trauma score were 13.6 and 13.4, respectively. Although statistically significant differences could not be demonstrated with this population size, trends were evident when the group was subdivided by age group (< or = 12 years and > 12 years of age). The younger group was more likely to require a thoracic operation after injury (35.3% v 23.5%), to have unstable vital signs (41.2% v 26.5%), to have a higher total abbreviated injury score (AIS) for the chest (4.8 v 4.0); however, their total extrathoracic AIS was lower (2.1 v 3.1). The younger group spend more time in the hospital and in the intensive care unit (7.6 v 4.6 days and 2.0 v 0.9 days, respectively). The younger children were more likely to have sustained injury by unintentional crossfire (35.3% v 14.7%) and were never injured by intentional assault (0% v 47.2%). Overall, 50% of the patients required surgery, including 93.8% of the patients who had unstable vital signs at the time of arrival. All six deaths (11.8%) owing to the thoracic injury occurred in patients who had mediastinal injury that required an emergency-department thoracotomy (EDT). The authors conclude that among this predominantly black male population, there are different trends with respect to treatment, circumstances, and pattern of injury between the younger and older subpopulations.

Angiotensin-converting enzyme inhibition limits dysfunction in adjacent noninfarcted regions during left ventricular remodeling

OBJECTIVES

We hypothesized that angiotensin-converting enzyme inhibitors would limit dysfunction in the first 8 weeks after transmural infarction in adjacent noninfarcted regions, as well as attenuate left ventricular remodeling.

BACKGROUND

Angiotensin-converting enzyme inhibition limits ventricular dilation and hypertrophy and improves survival after anterior infarction, but its effect on regional function during remodeling is not well characterized.

METHODS

Thirteen sheep underwent coronary ligation to create an anteroapical infarction. At postinfarction day 2, eight sheep were randomized to therapy with the angiotensin-converting enzyme inhibitor ramipril, and five sheep received no therapy. Animals were studied with magnetic resonance myocardial tagging before and 8 weeks after infarction. Left ventricular volume, mass and ejection fraction were measured, as were changes in percent circumferential shortening within the subendocardium and subepicardium of infarcted and noninfarcted myocardium, both adjacent to and remote from the infarction.

RESULTS

Angiotensin-converting enzyme inhibition limited the increase in end-diastolic volume from a mean (+/- SD) of +1.5 +/- 0.7 ml/kg in control animals to +0.5 +/- 0.8 ml/kg in the treated group (p < 0.04). Segmental function within infarcted and remote noninfarcted tissue did not differ between groups. However, angiotensin-converting enzyme inhibition limited the decline in function in the adjacent noninfarcted region 8 weeks after infarction. Percent circumferential shortening in the subendocardium decreased by -13 +/- 5% in the control group compared with -5 +/- 5% in the treated group (p < 0.03).

CONCLUSIONS

In concert with a reduction in left ventricular remodeling after anterior infarction, angiotensin-converting enzyme inhibition limits the decline in function in the adjacent noninfarcted region. Dysfunction in adjacent noninfarcted regions may be an important determinant of left ventricular remodeling after infarction.

Awareness of memory loss after severe closed-head injury

This study investigated unawareness of memory loss in 48 severe closed-head injury patients. Awareness was measured as the correlation of patients' memory test scores with their self-ratings on the Everyday Memory Questionnaire (EMQ). Patients who endorsed atypical memory failures on the EMQ were classified as invalid responders. Invalid responders had poorer memory test performances and a higher rate of focal left hemispheric lesions, but did not report greater emotional symptoms. Correlations between memory test scores and memory self-ratings were weaker among invalid responders or those classified as depressed, and stronger among valid responders, especially those classified as non-depressed. The results indicate that the validity of memory self-reports is influenced by both neurogenic and psychogenic factors, and suggest that invalid responding is an important problem with self-reports by brain-injured patients.

Large animal model of ischemic mitral regurgitation

A large animal model of ischemic mitral regurgitation (MR) that resembles the multiple presentations of the human disease was developed in sheep. In 76 sheep hearts, the anatomy of the coronary arterial circulation was determined by observation and polymer casts. Two variations, types A and B, which differed by the vessel that supplied the left ventricular apex, were found. In all hearts, the circumflex coronary artery has three marginal branches and terminates in the posterior descending coronary artery. The amount and location of left ventricular (LV) mass supplied by each marginal circumflex branch was determined by dye injection and planimetry. In type A hearts, ligation of the first and second marginal branches infarcts 23% +/- 3.0% of the LV mass, does not infarct either papillary muscle, significantly (p < 0.001) increases LV cavity size 48% at the high papillary muscle level by 8 weeks, and does not cause MR. Ligation of the second and third marginal branches infarcts 21.4% +/- 4.0% of the LV mass, includes the posterior papillary muscle, significantly increases (p < 0.001) LV cavity size 75%, and causes severe MR by 8 weeks. Ligation of the second and third marginal branches and the posterior descending coronary artery infarcts 35% to 40% of the LV mass, increases LV cavity size 39% within 1 hour, and causes massive MR. After moderate (21% to 23%) LV infarction, development of ischemic MR requires both LV dilatation and posterior papillary muscle infarction; neither condition alone produces MR. Large posterior wall infarctions (35% to 40%) that include the posterior papillary muscle produce immediate, severe MR.(ABSTRACT TRUNCATED AT 250 WORDS)

Socioeconomic impact of gunshot wounds in an urban pediatric population

Trauma continues to be a leading cause of death and disability in the pediatric population. We retrospectively reviewed 139 cases of gunshot wounds (GSW) in patients under 17 years of age, who were admitted to Children's Hospital of Philadelphia from January 1, 1986 to June 30, 1992. There were 4,587 trauma admissions during this period, and the proportion of GSW victims increased from 1.02% in the first 12 months to 5.6% in the final 12 months of the study (P < .001). The average age of patients treated for GSW was 11.5 +/- 4.4 years. Eighty percent were male, 88.5% were black, and the average hospital stay was 5.6 +/- 8.5 days. In this group, the mean trauma score was 14.5 (range, 1 to 16), and the mean injury severity score was 9.8 (range, 1 to 75). There were 11 deaths (7.9%), with head injuries the most common cause (20.6%). The predicted probability of survival for the patients who died was 0.31, compared with 0.97 for those who survived. The extremities were the most commonly injured area, (41%) but GSW to the abdomen were the most likely to require operative intervention (85.7%). The shootings were considered intentional in only 26.6%), and the assailant was known in 32.4% of instances. Of the teenage patients tested for substance abuse, results were positive for 36%. The majority of GSW victims (68%) had public or no insurance. In contrast, of the overall hospital population, only 40% had public or no insurance. Over the period of study, the hospital lost an estimated $1.63 million in the care of these patients.(ABSTRACT TRUNCATED AT 250 WORDS)

Home parenteral nutrition after near total enterectomy

Six patients with extreme short bowel syndrome (4.2 +/- 4.9 cm of residual small bowel) were provided home parenteral nutrition (HPN) for 14,397 days. The average age at onset of HPN was 38 years (18-64 years). Patients maintained body weight at 97% of ideal (86-112%) with mean serum albumin of 3.7 +/- 0.6 g/dL (normal 3.5-5.8 g/dL), serum transferrin of 341 +/- 104 mg/dL (normal 200-400 mg/dL), and mean serum pre-albumin of 27.5 +/- 12.6 mg/dL (normal 16.6-43 mg/dL). Hospital admission for HPN-related complications was required 10.3 times/patient for a total of 864 hospital days and was catheter related in 71% of episodes. Catheter sepsis occurred once per 436 catheter days and required catheter removal in only 33% of instances. Five patients were able to resume an oral diet, five had returned to work or school, and three were married with family. HPN allowed return to a quality productive life with acceptable morbidity following catastrophic massive bowel resection.

Pathogenesis of ischemic mitral insufficiency

We developed a new animal model of ischemic mitral insufficiency in sheep and used it to test the hypothesis that the combination of posterior papillary muscle infarction and left ventricular dilatation was required to produce mitral regurgitation after acute inferior myocardial infarction of moderate size. In 12 sheep, ligation of the first two circumflex marginal coronary arteries infarcted 23% of the left ventricular mass, increased left ventricular cavitary area from 13.2 +/- 1.2 cm2 to 20.0 +/- 2.7 cm2 by 8 weeks and did not produce ischemic mitral regurgitation. In 13 sheep, ligation of the second and third circumflex marginal arteries infarcted 21% of the left ventricular mass and, in 11 of these sheep, the posterior papillary muscular mass as well. When the papillary muscle was included, this infarction produced progressively severe mitral regurgitation over 8 weeks, as left ventricular cavitary area increased from 12.5 +/- 2.6 cm2 to 22.8 +/- 3.8 cm2. We conclude that neither posterior papillary muscle infarction nor left ventricular dilatation alone produces ischemic mitral regurgitation after moderate-sized inferior wall infarction, but that the combination does.