CT scans for pediatric injury in a middle-income country trauma center: Are we repeating past mistakes?

PURPOSE: Sustained efforts in high-income countries have decreased the rate of unnecessary computed tomography (CT) among children, aiming to minimize radiation exposure. There are little data regarding CT use for pediatric trauma in low- and middle-income countries. We aimed to assess the pattern and utility of CT performed during evaluation of trauma patients presenting to a middle-income country (MIC) trauma center. METHODS: We reviewed pediatric (age<18) trauma admissions at a single tertiary referral center in South Africa. Patient demographics, injury details, surgical intervention(s), and mortality were abstracted from the medical record. CT indications, results, and necessity were determined by review. RESULTS: Of 1,630 children admitted to the trauma center, 826 (51%) had CT imaging. Children undergoing imaging were younger (median age 11 [IQR: 6, 16] vs 13 [IQR: 7, 17]) and had higher median ISS [9 [IQR: 4, 13] vs 4 [2, 9]) compared to those without imaging (both p<0.001). Overall, 1,224 scans were performed with normal findings in 609 (50%). A median of 1 scan was performed per patient (range: 1-5). The most common location was CT head (n = 695, 57%). Among patients with positive findings on CT head (n = 443), 31 (7%) underwent either intracranial pressure monitoring or surgery. CT of the cervical spine had positive findings in 12 (7%) with no patients undergoing spine surgery. Of 173 patients with abdominal CT imaging, 83 (48%) had abnormal findings and 18 (10%) required operative exploration. Thirteen (16%) patients with abnormal findings on abdominal CT had exploratory laparotomy. Of 111 children undergoing whole body CT, 8 (7%) underwent thoracic and/or abdominal operations. CONCLUSION: Use of CT during evaluation of pediatric trauma is common in an MIC center. A high rate of normal findings and low rates of intervention following head, cervical spine, and abdominal CT suggest potential overuse of this resource. LEVEL OF EVIDENCE: Level III.

Mob Justice in South Africa: A comparison of blunt trauma secondary to community and non-community assaults.

BACKGROUND: The phenomenon of extrajudicial "mob justice" and community assault (CA) has been documented in news reports and anecdotes from a number of low- and middle-income countries, but there is little literature on its burden on trauma systems. This study reviews a single center's management of CA victims and compares the spectrum of injuries seen following mob assault with those sustained via other forms of interpersonal violence (IPV). METHODS: Clinical data, injury details, and mortality among injured patients (age≥18) hospitalized in a South African tertiary referral center from 2012-2018 were abstracted. Patients with penetrating injury or missing ISS were excluded. CA was determined at time of admission by either self-designation or by patients' escorts. Univariate analyses compared the presentation and outcomes for CAs and non-CAs. RESULTS: Overall, CA constituted 5% of total trauma admissions and 8% of IPV-related admissions during the study period. Of 1,323 incidents of blunt injury following IPV, 239 (18%) were CAs. One in two CA victims (n=119, 50%) were struck by an identifiable weapon. Patients injured in CA were more frequently male (97% vs 85%), presented with ISS>15 (28% vs 21%), and had a shock index>0.9 (25% vs 19%) compared to non-CA (all p<0.001). Rates of operative intervention, ICU admission, and mortality did not differ (all p>0.05). CAs were more likely to be complicated by acute kidney injury (9% vs 1%, p<0.001) but less likely to involve neurologic complications (3% vs 10%. P<0.001) compared to non-CAs. Acute kidney injury in CA showed a pattern of significant musculocutaneous injury with rhabdomyolysis. CONCLUSION: CA contributes considerably to the high rates of IPV in a single South African center. Victims of such assaults sustain more severe injury with unique mechanisms and subsequent complications. This evidence supports the need to strengthen local governance and improve law enforcement efforts to prevent such violence.

Use of vital signs in predicting surgical intervention in a South African population: A cross-sectional study.

BACKGROUND: While vital signs are widely obtained for trauma patients around the world, the association of these signs with surgical intervention has yet to be defined. Early detection of trauma patients requiring surgery is essential to timely intervention and reduction of morbidity and mortality. OBJECTIVE: The aim of this study was to determine the association of vital signs with surgical intervention in a South African patient population. METHODS: This retrospective cohort included 7857 trauma patients admitted at Grey's Hospital in Pietermaritzburg, South Africa over a five-year period December 2012-April 2018. Exclusion criteria included missing key data points. Variables for analysis included sex, mechanism of injury, admission Glasgow Coma Scale (GCS), systolic blood pressure, diastolic blood pressure, temperature, heart rate, and respiratory rate. Surgical intervention was defined by the need for treatment requiring time in the operating room. Data were analyzed using a univariate and multivariate logistic regression to determine an association between admission vital signs and surgical intervention and was compared to the association of the Revised Trauma Score to surgical intervention. RESULTS: Of the 8722 trauma patient records available, exclusion of patients with incomplete data resulted in 7857 patient records available for analysis. Two thousand two hundred and ninety-six (29.2%) patients required surgical intervention in the operating room. Multivariate analysis revealed that male sex [odds ratio (OR) 1.25, 95% confidence interval (CI) 1.06-1.48], stab wound (OR 3.42, CI 2.99-3.09), gunshot wound (OR 4.27, CI 3.58-5.09), systolic hypotension (OR 1.81, CI 1.32-2.48), hypothermia (OR 1.77, CI 1.34-2.34), tachycardia (OR 1.84, CI 1.61-2.10), and tachypnea (OR 1.26, CI 1.08-1.45) were associated with an increased likelihood of surgical intervention. CONCLUSIONS: In this cohort of patients, the need for surgical intervention was best predicted by penetrating mechanisms of injury, tachycardia, and systolic hypotension. These data show that rapid and focused patient assessments should be used to triage patients for emergency surgery to avoid delays.

Comparison of the Pediatric Resuscitation and Trauma Outcome (PRESTO) Model and Pediatric Trauma Scoring Systems in a Middle-Income Country.

BACKGROUND: The pediatric resuscitation and trauma outcome (PRESTO) model was developed to aid comparisons of risk-adjusted mortality after injury in low- and middle-income countries (LMICs). We sought to validate PRESTO using data from a middle-income country (MIC) trauma registry and compare its performance to the Pediatric Trauma Score (PTS), Revised Trauma Score, and pediatric age-adjusted shock index (SIPA). METHODS: We included children (age < 15 years) admitted to a single trauma center in South Africa from December 2012 to January 2019. We excluded patients missing variables necessary for the PRESTO model-age, systolic blood pressure, pulse, oxygen saturation, neurologic status, and airway support. Trauma scores were assigned retrospectively. PRESTO's previously high-income country (HIC)-validated optimal threshold was compared to MIC-validated threshold using area under the receiver operating characteristic curves (AUROC). Prediction of in-hospital death using trauma scoring systems was compared using ROC analysis. RESULTS: Of 1160 injured children, 988 (85%) had complete data for calculation of PRESTO. Median age was 7 (IQR: 4, 11), and 67% were male. Mortality was 2% (n = 23). Mean predicted mortality was 0.5% (range 0-25.7%, AUROC 0.93). Using the HIC-validated threshold, PRESTO had a sensitivity of 26.1% and a specificity of 99.7%. The MIC threshold showed a sensitivity of 82.6% and specificity of 89.4%. The MIC threshold yielded superior discrimination (AUROC 0.86 [CI 0.78, 0.94]) compared to the previously established HIC threshold (0.63 [CI 0.54, 0.72], p < 0.0001). PRESTO showed superior prediction of in-hospital death compared to PTS and SIPA (all p < 0.01). CONCLUSION: PRESTO can be applied in MIC settings and discriminates between children at risk for in-hospital death following trauma. Further research should clarify optimal decision thresholds for quality improvement and benchmarking in LMIC settings.

The search for a simple injury score to reliably discriminate the risk of in-hospital mortality in South Africa.

BACKGROUND: The Injury Severity Score and Trauma and Injury Severity Score are used commonly to quantify the severity of injury, but they require comprehensive data collection that is impractical in many low- and middle-income countries . We sought to develop an injury score that is more feasible to implement in low- and middle-income countries with discrimination similar to the Injury Severity Score and the Trauma and Injury Severity Score. METHODS: Clinical data from KwaZulu-Natal, South Africa were used to compare the discrimination of the Injury Severity Score and the Trauma and Injury Severity Score with that of the 5, simple injury scores that rely primarily on physiologic data: Revised Trauma Score for Triage, "Mechanism, Glasgow Coma Scale, Age, Pressure" Score, Kampala Trauma Score, modified Kampala Trauma Score, and "Reversed Shock Index Multiplied by Glasgow Coma Scale" Score. RESULTS: Data for 3,991 patients were analyzed. The Trauma and Injury Severity Score, the Injury Severity Score, and Kampala Trauma Score had similar discrimination (area under the receiver operating curve 0.85, 0.84, and 0.84, respectively). The simple injury scores demonstrated worse discrimination among patients presenting more than 6 hours postinjury, although Kampala Trauma Score maintained the best discrimination of the simple injury scores. CONCLUSION: In this patient population, Kampala Trauma Score demonstrated discrimination similar to the Injury Severity Score and the Trauma and Injury Severity Score and may be useful to quantify the severity of injury when calculation of the Injury Severity Score or the Trauma and Injury Severity Score is not feasible. Delay in presentation can degrade the discrimination of simple injury scores that rely primarily on physiologic data.

Predicting the Outcome of Non-operative Management of Splenic Trauma in South Africa.

INTRODUCTION: We aimed to expand on the global surgical discussion around splenic trauma in order to understand locally and clinically relevant factors for operative (OP) and non-operative management (NOM) of splenic trauma in a South African setting. METHODS: A retrospective cohort study was performed using 2013-2017 data from the Pietermaritzburg Metropolitan Trauma Service. All adult patients (≥15 years) were included. Those managed with OP or NOM for splenic trauma were identified and analyzed descriptively. Multiple logistic regression analysis identified patients and clinical factors associated with management type. RESULTS: There were 127 patients with splenic injury. Median age was 29 [19-35] years with 42 (33%) women and 85 (67%) men. Blunt injuries occurred in the majority (81, 64%). Organ Injury Scale (OIS) grades included I (25, 20%), II (43, 34%), III (36, 28%), IV (15, 11%), and V (8, 6%). Nine patients expired. On univariate analysis, increasing OIS was associated with OP management, need for intensive care unit (ICU) admission, and hospital and ICU duration of stay, but not mortality. In patients with a delayed compared to early presentation, ICU utilization (62% vs. 36%, p = 0.008) and mortality (14% vs. 4%, p = 0.03) were increased. After adjusting for age, sex, presence of shock, and splenic OIS, penetrating trauma (adjusted odds ratio, 5.7; 95%CI, 1.7-9.8) and admission lactate concentration (adjusted odds ratio, 1.4; 95%CI 1.1-1.9) were significantly associated with OP compared to NOM (p = 0.002; area under the curve 0.81). CONCLUSIONS: We have identified injury mechanism and admission lactate as factors predictive of OP in South African patients with splenic trauma. Timely presentation to definitive care affects both ICU duration of stay and mortality outcomes. Future global surgical efforts may focus on expanding non-operative management protocols and improving pre-hospital care in patients with splenic trauma.

Evolving Indications for Lower Limb Amputations in South Africa Offer Opportunities for Health System Improvement.

BACKGROUND: Rapid urbanization and westernization have precipitated dramatic changes in the profile and prevalence of surgical diseases in sub-Saharan Africa. Disease of lifestyle is now common. We aimed to review our experience with lower-limb amputations at our surgical service in South Africa. METHODS: A single-center retrospective review of a prospectively collected database was performed of all patients who underwent a lower limb amputation. Inferential and descriptive statistics were performed. Patient demographics, indication, type of amputation, and management were reviewed. The primary outcome was 30-day in-patient mortality rate. RESULTS: Over a 5-year period (2013-2018), 348 patients underwent lower limb amputations. The median age was 61.5 years. 53.7% were diabetic and 56.3% were hypertensive. 53.2% had associated peripheral vascular disease and 8% preexisting cardiac disease. 30.7% smoked. Guillotine below-knee amputation was frequently performed (44.5% of amputations). 16.1% of these patients required a further operation. The in-hospital mortality rate was 8%. Underlying renal disease was an independent risk factor for mortality (p = 0.004). CONCLUSION: Currently, the most common indications for LLA in South Africa are diabetes mellitus and atherosclerosis. This reflects the changing pattern of disease in the country. There is a major problem with access to health care in rural areas in South Africa with significant delays in getting patients to tertiary units for evaluation by specialists. Foot care and prevention at a primary health care level is also lacking. Global improvements in the healthcare system are needed to improve LLA rates in South Africa.

Utilization of age-adjusted shock index in a resource-strained setting.

BACKGROUND: Identification of injury severity and appropriate triage are critical to effective surgical care, especially where medical and surgical resources are strained. We hypothesized that pediatric age-adjusted shock index (SIPA) would outperform traditional shock index (SI) in a middle-income country (MIC) setting. METHODS: Injured children hospitalized in two trauma centers (South Africa and the United States) from 2012 to 2017 were reviewed. Maximum heart rate and minimum systolic blood pressure defined SI. SI > 0.9 defined elevation. SIPA elevation was based on SI stratified by age: 1-6 years (SI > 1.22), 7-12 years (SI > 1.0), and 13-17 years (SI > 0.9). SI and SIPA were compared using univariate analyses and area under the receiver operating characteristic curves (AUROC). RESULTS: 1648 patients (741 MIC and 907 high-income country (HIC)) were evaluated with a median [IQR] age of 11 [6-15] years. SI was elevated in 377 (51%) MIC children, whereas SIPA was elevated in 248 (34%). In both the HIC and MIC, elevated SIPA was more associated with ISS ≥ 25, ICU admission, and mortality. In MIC patients specifically, elevated SIPA improved discrimination for in-hospital mortality (AUROC 0.66 vs AUROC 0.57, p < 0.01). CONCLUSION: In a multinational cohort including MIC patients, SIPA facilitated identification of injured children with altered physiology, reflecting greater injury severity and poorer outcomes. Use of SIPA has the potential for more effective resource utilization in MICs. LEVEL OF EVIDENCE: Level III.

Trauma registry data as a tool for comparison of practice patterns and outcomes between low- and middle-income and high-income healthcare settings.

PURPOSE: There is a lack of data-driven, risk-adjusted mortality estimates for injured children outside of high-income countries (HIC). To inform injury prevention and quality improvement efforts, an upper middle-income country (UMIC) pediatric trauma registry was compared to that of a HIC. METHODS: Clinical data, injury details, and mortality of injured children (< 18 years) hospitalized in two centers (USA and South African (SA)) from 2013 to 2017 were abstracted. Univariate and multivariable analyses evaluated risk of mortality and were expressed as odds ratios (OR) with 95% confidence intervals (CI). RESULTS: Of 2089 patients, SA patients had prolonged transfer times (21.1 vs 3.4 h) and were more likely referred (78.2% vs 53.9%; both p < 0.001). Penetrating injuries were more frequent in SA (23.2% vs 7.4%, p < 0.001); injury severity (9 vs 4) and shock index (0.90 vs 0.80) were greater (both p < 0.001). SA utilized cross-sectional imaging more frequently (66.4% vs 37.3%, p < 0.001). In-hospital mortality was similar (1.9% SA, 1.3% USA, p = 0.31). Upon multivariable analysis, ISS > 25 [210.50 (66.0-671.0)] and penetrating injury [5.5 (1.3-23.3)] were associated with mortality, while institution [1.7 (0.7-4.2)] was not. CONCLUSIONS: Despite transfer time, the centers demonstrated comparable survival rates. Comparison of registry data can alert clinicians to problematic practice patterns, assisting initiatives to improve trauma systems.

Surgical Considerations for Pediatric Snake Bites in Low- and Middle-Income Countries.

BACKGROUND: Snake envenomation is associated with major morbidity especially in low- and middle-income countries and may require fasciotomy. We determined patient factors associated with the need for fasciotomy after venomous snake bites in children located in KwaZulu-Natal, South Africa. METHODS: Single institutional review of historical data (2012-2017) for children (<18 years) sustaining snake envenomation was performed. Clinical data, management, and outcomes were abstracted. Syndromes after snake bite were classified according to Blaylock nomenclature: progressive painful swelling (PPS), progressive weakness (PW), or bleeding (B), as it is difficult to reliably identify the species of snake after a bite. Comparative and multivariable analyses to determine factors associated with fasciotomy were performed. RESULTS: There were 72 children; mean age was 7 (±3) years, 59% male. Feet were most commonly affected (n = 27, 38%) followed by legs (n = 18, 25%). Syndromes (according to Blaylock) included PPS (n = 63, 88%), PW (n = 5, 7%), and B (n = 4, 5%). Eighteen patients underwent fasciotomy, and one required above knee amputation. Nine patients received anti-venom. Few patients (15%) received prophylactic beta-lactam antibiotics. Hemoglobin < 11 mg/dL, leukocytosis, INR >1.2, and age-adjusted shock index were associated with fasciotomy. On regression, age-adjusted shock index and hemoglobin concentration < 11 mg/dL, presentation >24 h after snake bite, and INR >1.2 were independently associated with fasciotomy. Model sensitivity was 0.89 and demonstrated good fit. CONCLUSIONS: Patient factors were associated with the fasciotomy. These factors, coupled with clinical examination, may identify those who need early operative intervention. Improving time to treatment and the appropriate administration of anti-venom will minimize the need for surgery. LEVEL OF EVIDENCE: III.

Laparotomy for organ evisceration from abdominal stab wounds: A South African experience.

INTRODUCTION: Organ evisceration following abdominal stab wound (SW) is currently considered as an absolute indication for mandatory laparotomy due to the high incidence of associated intra-abdominal injuries, but literature describing the spectrum of organ injury encountered is limited. MATERIALS AND METHODS: We reviewed our experience of 301 consecutive patients who were subjected to mandatory laparotomy over an eight-year period at a major trauma centre in South Africa. RESULTS: Of the 301 patients with organ evisceration, 92% were male (mean age: 28 years). Ninety per cent (270/301) of the laparotomies were positive (85% (229/270) therapeutic, 15% (41/270) non-therapeutic). The frequencies of eviscerated organs were small bowel (70%), large bowel (26%), and stomach 3%. Three (1%) patients had combined evisceration of more than one of the above organs. The most commonly injured organs were small bowel and large bowel. The mean length of hospital stay was nine days. Seven patients required intensive care admission. The morbidity rate was 21% and mortality was 2%. CONCLUSIONS: The spectrum of injury associated with abdominal SW with organ evisceration is similar to smaller published series. Multiple organ injuries are common. The most commonly eviscerated organs were small bowel, large bowel and stomach, while the most commonly injured organs were small bowel and large bowel.

Defining an intra-operative blunt mesenteric injury grading system and its use as a tool for surgical-decision making.

BACKGROUND: The mesentery may be injured in trauma and few grading systems describe mesenteric injury severity. We aimed to develop and validate an intra-operative mesenteric injury grading system. METHODS: A modified Delphi technique was used to generate an intraoperative grading system for blunt mesenteric injury called the mesenteric injury score (MIS). We performed a retrospective review (2010-2016) of patients >15 years old with blunt abdominal trauma. Patient demographics, injury severity score (ISS) and mechanism, clinical, operative, and outcome data were abstracted. The intraoperative grading system was used to describe patient outcomes including duration of stay and management approach. We compared the correlation of abdominal abbreviated injury score, Blunt Injury Prediction Score (BIPS) and the MIS with clinical outcomes using Spearman's rho. RESULTS: There were fifty-one patients of which 86% were male. Injury mechanisms included motor vehicle accidents (n = 37, 73%), pedestrian vehicle accidents (n = 7, 13%), assaults (n = 4, 8%), falls (n = 2, 4%), and a single airplane crash (2%). Median [IQR] ISS was 16 [10-25] and GCS at hospital admission was 15 [15-15]. The median [IQR] international normalized ratio was 1.2 [1.1-1.5], lactate was 2.7 [1.7-4.9], and hemoglobin was 11.4 [8.6-12.2]. The distributions of MIS included Grade I (3, 5%), Grade II (10, 20%), Grade III (10, 20%), Grade IV, 5 (10%), and Grade V (23, 45%). Increasing mesenteric injury grade was associated with longer duration of stay, need for bowel resection, and damage control laparotomy. CONCLUSIONS: We developed an intra-operative mesenteric injury grading system (MIS) and provided an initial retrospective validation using a series of patients with blunt abdominal trauma. The proposed MIS corresponded with both the AIS and the BIPS. Future study comparing cross sectional imaging and operative findings based on MIS criteria is needed.

Acute Kidney Injury on Presentation to a Major Trauma Service is Associated with Poor Outcomes.

BACKGROUND: This study sought to describe the burden of disease of acute kidney injury (AKI) among adult South African trauma patients who presented to a tertiary level trauma service. METHODS: The trauma database was interrogated for the period from December 2012 to July 2017. All patients over the age of 18 y, who were admitted following trauma, were included. Outcome data were reviewed. This included in-hospital mortality, need for intensive care unit admission, and length of stay. AKI was defined according to the latest Kidney Disease Improving Global Outcomes guidelines using the presentation serum creatinine. RESULTS: A total of 7613 patients were admitted for trauma over the period under review. Four thousand two hundred sixty-six patients were suitable for analysis. A total of 238 (5.6%) patients presented with AKI, 149 (62.6%) had stage 1 AKI, 40 (16.8%) had stage 2 AKI, and 49 (20.6%) had stage 3 AKI. There was a higher incidence of AKI in patients with blunt trauma. The length of stay, need for intensive care unit admission, and mortality were significantly higher in patients presenting with AKI than in those who did not present with AKI. There were 172 deaths (4.0%). The patients who died were older and had significantly higher Injury Severity Score than survivors. They were more acidotic on presentation, had lower Glasgow Coma Scale, and were more likely to be hypotensive on presentation. They also were significantly more likely to have AKI on presentation. (30.2% versus 5.6% P < 0.001). AKI on presentation was an independent risk factor for mortality (odds ratio 3.038 95% confidence interval 1.260-7.325). CONCLUSIONS: AKI is common in patients presenting to our center with acute trauma. The presence of AKI is associated with increased morbidity and mortality. Efforts must be directed to improving recognition of at-risk patients. Prompt referral and adequate resuscitation of trauma patients before transfer must be prioritized.

Validation of the AAST EGS grading system for perforated peptic ulcer disease.

INTRODUCTION: Perforated peptic ulcer disease (PPUD) including both duodenl and gastric ulcers is a severe disease and outcomes are influenced by comorbidities and physiology. We validated the AAST EGS grading system at two diverse centers (Mayo Clinic, USA and Pietermaritzburg, South Africa). METHODS: Dual-center review of historic data (2010-2016) of adults with PPUD was performed. Preoperative, procedural, and postoperative data were abstracted. ASA, Boey, PULP and AAST EGS grades were generated. Comparative, multivariable, and pairwise analyses were performed. RESULTS: There were 306 patients, 42% female with a mean (±SD) age of 56 ±20 years. Overall, the patints were categorized into the following AAST EGS grades: I (30, 10%), II (38, 12%), III (104, 34%), IV (76, 2e%), V (58, 18.9%). Initial management included: midline laparotomy (51%, n=157), laparoscopy (18%, n=58), laparoscopy converted to laparotomy (1%, n=3), and endoscopy (30%, n=88). Duration of stay increased with AAST EGS grade. In United States cohort, factors predictive for 30-day mortality included AAST EGS grade and patient comorbidity status. The AAST EGS grade was comparable to other scoring systems (Boey, PULP, and ASA). CONCLUSIONS: Differences exist between centers for management of PPUD and their outcomes; however, the AAST EGS grade can be utilized to stratify thedisease severity of the patient and this demonstrates initial construct validity in a United States but not in a South African population.

Validation of the American Association for the Surgery of Trauma's emergency general surgery breast infection grading system.

BACKGROUND: The American Association for the Surgery of Trauma (AAST) developed emergency general surgery (EGS) grading systems for multiple diseases to standardize classification of disease severity. The grading system for breast infections has not been validated. We aimed to validate the AAST breast infection grading system. METHODS: Multi-institutional retrospective review of all adult patients with a breast infection diagnosis at Mayo Clinic Rochester 1/2015-10/2015 and Pietermaritzburg South African Hospital 1/2010-4/2016 was performed. AAST EGS grades were assigned by two independent reviewers. Inter-rater reliability was measured using the agreement statistic (kappa). Final AAST grade was correlated with patient and treatment factors using Pearson's correlation coefficient. RESULTS: Two hundred twenty-five patients were identified: grade I (n = 152, 67.6%), II (n = 44, 19.6%), III (n = 25, 11.1%), IV (n = 0, 0.0%), and V (n = 4, 1.8%). At Mayo Clinic Rochester, AAST grades ranged from I-III. The kappa was 1.0, demonstrating 100% agreement between reviewers. Within the South African patients, grades included II, III, and V, with a kappa of 0.34, due to issues of the grading system application to this patient population. Treatment received correlated with AAST grade; less severe breast infections (grade I-II) received more oral antibiotics (correlation [-0.23, P = 0.0004]), however, higher AAST grades (III) received more intravenous antibiotics (correlation 0.29, P <0.0001). CONCLUSIONS: The AAST EGS breast infection grading system demonstrates reliability and ease for disease classification, and correlates with required treatment, in patients presenting with low-to-moderate severity infections at an academic medical center; however, it needs further refinement before being applicable to patients with more severe disease presenting for treatment in low-/middle-income countries.

Pediatric Appendicitis Severity in KwaZulu-Natal, South Africa: A Retrospective Cohort Analysis.

BACKGROUND: Acute appendicitis is a common pediatric surgical emergency; however, there are few grading systems to assign disease severity. The American Association for the Surgery of Trauma (AAST) recently developed a grading system for a variety of emergency surgical conditions, including appendicitis. The severity of acute appendicitis in younger patients in KwaZulu-Natal (South Africa) is unknown. We aimed to describe the disease severity in this patient population using the AAST grading system hypothesizing that the AAST grade would correlate with morbidity, management type, and duration of stay. MATERIALS: Single institutional review of patients <18 years old with a final diagnosis of acute appendicitis during 2010-2016 in KwaZulu-Natal, South Africa, was performed. Demographics, physiologic and symptom data, procedural details, postoperative complications, and Clavien-Dindo classification were abstracted. AAST grades were generated based on intraoperative findings. Summary, univariate, and nominal logistic regression analyses were performed to compare AAST grade and outcomes. RESULTS: A total of 401 patients were identified with median [IQR] age of 11 [5-13], 65% male. Appendectomy was performed in all patients; 2.4% laparoscopic, 37.6% limited incision, and 60% midline laparotomy. Complications occurred in 41.6%, most commonly unplanned relaparotomy (22.4%), surgical site infection (8.9%), pneumonia (7.2%), and acute renal failure (2.9%). Complication rate and median length of stay increased with greater AAST grade (all p < 0.001). AAST grade was independently associated with increased risk of complications. CONCLUSION: Pediatric appendicitis is a morbid disease in a developing middle-income country. The AAST grading system is generalizable and accurately corresponds with management strategies as well as key clinical outcomes. LEVEL OF EVIDENCE: Retrospective study, Level IV. STUDY TYPE: Retrospective single institutional study.

Routine cervical spine immobilisation is unnecessary in patients with isolated cerebral gunshot wounds: A South African experience.

OBJECTIVE: Routine immobilisation of the cervical spine in trauma has been a long established practice. Very little is known in regard to its appropriateness in the specific setting of isolated traumatic brain injury secondary to gunshot wounds (GSWs). METHODS: A retrospective study was conducted over a 5 year period (January 2010 to December 2014) at the Pietermaritzburg Metropolitan Trauma Service, Pietermaritzburg, South Africa in order to determine the actual incidence of concomitant cervical spine injury (CSI) in the setting of isolated cerebral GSWs. RESULTS: During the 5 year study period, 102 patients were included. Ninety-two per cent (94/102) were male and the mean age was 29 years. Ninety-eight per cent of the injuries were secondary to low velocity GSWs. Twenty-seven (26%) patients had cervical collar placed by the Emergency Medical Service. The remaining 75 patients had their cervical collar placed in the resuscitation room. Fifty-five (54%) patients had a Glasgow Coma Scale (GCS) of 15 and underwent plain radiography, all of which were normal. Clearance of cervical spine based on normal radiography combined with clinical assessment was achieved in all 55 (100%) patients. The remaining 47 patients whose GCS was <15 all underwent a computed tomography (CT) scan of their cervical spine and brain. All 47 CT scans of the cervical spine were normal and there was no detectable bone or soft tissue injury noted. CONCLUSION: Patients who sustain an isolated low velocity cerebral GSW are highly unlikely to have concomitant CSI. Routine cervical spine immobilisation is unnecessary, and efforts should be directed at management strategies aiming to prevent secondary brain injury. Further studies are required to address the issue in the setting of high velocity GSWs.

Measuring Anatomic Severity in Pediatric Appendicitis: Validation of the American Association for the Surgery of Trauma Appendicitis Severity Grade.

OBJECTIVE: To assess whether the American Association for the Surgery of Trauma (AAST) grading system accurately corresponds with appendicitis outcomes in a US pediatric population. STUDY DESIGN: This single-institution retrospective review included patients <18 years of age (n = 331) who underwent appendectomy for acute appendicitis from 2008 to 2012. Demographic, clinical, procedural, and follow-up data (primary outcome was measured as Clavien-Dindo grade of complication severity) were abstracted. AAST grades were generated based on intraoperative findings. Summary, univariate, and multivariable regression analyses were performed to compare AAST grade and outcomes. RESULTS: Overall, 331 patients (46% female) were identified with a median age of 12 (IQR, 8-15) years. Appendectomy was laparoscopic in 90% and open in 10%. AAST grades included: Normal (n = 13, 4%), I (n = 152, 46%), II (n = 90, 27%), III (n = 43, 13%), IV (n = 24 7.3%), and V (n = 9 2.7%). Increased AAST grade was associated with increased Clavien-Dindo severity, P =.001. The overall complication rate was 13.6% and was comprised by superficial surgical site infection (n = 13, 3.9%), organ space infection (n = 15, 4.5%), and readmission (n = 17, 5.1%). Median duration of stay increased with AAST grade (P < .0001). Nominal logistic regression identified the following as predictors of any complication (P < .05): AAST grade and febrile temperature at admission. CONCLUSIONS: The AAST appendicitis grading system is valid in a single-institution pediatric population. Increasing AAST grade incrementally corresponds with patient outcomes including increased risk of complications and severity of complications. Determination of the generalizability of this grading system is required.

Appendicitis: Rural Patient Status is Associated with Increased Duration of Prehospital Symptoms and Worse Outcomes in High- and Low-Middle-Income Countries.

INTRODUCTION: Appendicitis is a significant economic and healthcare burden in low-, middle-, and high-income countries. We aimed to determine whether urban and rural patient status would affect outcomes in appendicitis in a combined population regardless of country of economic status. We hypothesize that patients from rural areas and both high- and low-middle-income countries would have disproportionate outcomes and duration of symptoms compared to their urban counterparts. METHODS: Adults (≥18 years) with appendicitis during 2010-2016 in South Africa and USA were reviewed using multi-institutional data. Baseline demographic, operative details, durations of stay, and complications (Clavien-Dindo index) were collected. AAST grades were assigned by two independent reviewers based on operative findings. Summary, univariate, and multivariable analyses of rural and urban patients in both countries were performed. RESULTS: There were 2602 patients with a median interquartile range [IQR] of 26 [18-40] years; 45% were female. Initial management included McBurney incisions (n = 458, 18%), laparotomy (n = 915, 35%), laparoscopic appendectomy (n = 1185, 45%), and laparoscopy converted to laparotomy (n = 44, 2%). Comparing rural versus urban patient status, there were increased overall median [IQR] AAST grades (3 [1-5] vs. 2 [1-3], p = 0.001), prehospital duration of symptoms (2 [1-5] vs. 2 [1-3], p = 0.001), complications (44.3 vs. 23%, p = 0.001), and need for temporary abdominal closure (20.3 vs. 6.9%, p = 0.001). CONCLUSION: Despite socioeconomic status and country of origin, patients from more rural environments demonstrate poorer outcomes notwithstanding significant differences in overall disease severity. The AAST grading system may serve a potential benchmark to recognize areas with disparate disease burdens. This information could be used for strategic improvements for surgeon placement and availability.

The effect of HIV status on clinical outcomes of surgical sepsis in KwaZulu-Natal Province, South Africa.

BACKGROUND: KwaZulu-Natal Province, South Africa (SA), has long been the epicentre of the HIV epidemic, but the impact of HIV co-infection on the clinical outcomes of emergency surgical patients with sepsis remains largely unknown. OBJECTIVE: To review our experience with the management of patients with HIV co-infection and to compare the disease spectrum and outcome with those without HIV infection. METHODS: A retrospective study was undertaken at the Pietermaritzburg Metropolitan Surgical Service (PMSS), SA over a 5-year period from January 2010 to December 2014. RESULTS: A total of 675 patients with a documented surgical source of sepsis were reviewed. Of these, 332 (49%) were male, and the mean age was 46 (standard deviation 19) years. HIV status was known in 237 (35%) patients, 146 (62%) were HIV-positive and the remaining 91 (38%) were HIV-negative. Other than tuberculosis of the abdomen being significantly more common in HIV-positive than HIV-negative patients (10% v. 2%, p=0.033), there were no differences in the spectrum of diseases between the two groups. There were no significant differences in overall morbidity or mortality. When adjusted for CD4 counts, the mortality in HIV-positive patients with a CD4 count <200 cells/µL was 60% (15/25) and in those with a CD4 count >200 cells/µL it was 2% (2/101) (p<0.001). CONCLUSION: The clinical presentation and the spectrum of surgical sepsis in patients with HIV co-infection were not markedly different to those in patients who were not HIV-infected. HIV-infected patients with a CD4 count <200 cells/µL had a significantly higher mortality. Management approaches should not differ based solely on the HIV status of patients with surgical sepsis.