64.10 Current Trends in FFP Transfusion & VTE Prophylaxis Following Hepatectomy: A Survey Analysis

Posted on December 22, 2014

J. N. Leal1, T. P. Kingham1, P. J. Allen1, R. P. DeMatteo1, W. R. Jarnagin1, M. I. D’Angelica1  1Memorial Sloan-Kettering Cancer Center,Surgery,New York, NY, USA

Introduction:

Following hepatectomy INR is often elevated and elicits concern regarding increased bleeding risk. This can lead to transfusion of fresh frozen plasma (FFP) and/or delay the institution of venous thromboembolism (VTE) prophylaxis.  INR, however, does not accurately reflect coagulation status following hepatectomy, and despite observed elevations, hypercoaguability and thrombosis are common. The clinical usefulness of the INR following hepatectomy is therefore questionable. The purpose of this study is to characterize current practice patterns regarding INR, FFP transfusion triggers and VTE prophylaxis following hepatectomy amongst a group of liver surgeons.

Methods:
A survey addressing FFP transfusion triggers and characteristics of VTE administration following hepatectomy was developed and distributed to the active membership of America’s Hepatopancreaticobiliary Association (AHPBA). Results were summarized for the group as a whole and then stratified into groups based on reported number of hepatectomies/year; Group A (≥50 hepatectomies/year) and Group B (< 50 hepatectomies/year), and responses compared.

Results:
Surveys were emailed to active members of the AHPBA (n=971). Overall 174(18%) surveys were completed. Post operative FFP transfusion rate was estimated to be <25% by the majority or respondents (149, 86%). The most commonly reported trigger for FFP transfusion was clinical evidence of bleeding (42%). However, over a third of participants (63, 36%) reported using isolated INR elevation as the sole trigger for FFP transfusion.  Amongst these respondents the level of INR triggering FFP transfusion varied considerably (range 1.5-2.0). VTE prophylaxis following liver resection was reported to be utilized by 94% of respondents, however, only 54% of gave it in accordance with current guidelines. Elevated INR was reported as the reason for not implementing VTE prophylaxis at all (8, 5%) or delaying implementation for >24 hrs (61, 35%). No differences between groups A (n=82) and B (n=90) in terms of type of transfusion trigger was observed (p=0.18). However, a trend towards use of a higher INR (>1.8) as the trigger point in group A compared to group B was observed (Figure 1, p=0.07). No differences between groups were observed with regards to type and timing of VTE prophylaxis.

Conclusion:

Despite recent evidence questioning the accuracy of INR following liver resection; its use as a trigger for transfusion and/or delay in VTE prophylaxis remains common. In centers where higher numbers of hepatectomies are performed it appears tolerance for higher INR prior to transfusion may exist.

 

64.11 Patterns of Care Among Patients Undergoing Hepatic Resection: A Query of the NSQIP Database

Posted on December 22, 2014

G. Spolverato1, A. Ejaz1, Y. Kim1, B. L. Hall2, K. Bilimoria3, M. Cohen4, C. Ko5, H. Pitt6, T. M. Pawlik1  2Washington University,Department Of Surgery,,St. Louis, MISSOURI, USA 3Northwestern University,Department Of Surgery,Chicago, ILLINOIS, USA 4American College Of Surgeons,Chicago, ILLINOIS, USA 5University Of California Los Angeles,Los Angeles, CALIFORNIA, USA 6Temple University Health Systems,Philadelphia, PENNSYLVANIA, USA 1Johns Hopkins University School Of Medicine,Baltimore, MD, USA

Introduction:  The American College of Surgeons recently added liver-specific variables to the National Surgical Quality Improvement Program (NSQIP).  We sought to utilize these variables to define patterns of care, as well as characterize peri-operative outcomes among patients undergoing hepatic resection (HR).

Methods:  The ACS-NSQIP database was queried for all patients undergoing HR between January 1, 2013 and October 15, 2013 (n=1,556).  Liver-specific variables were summarized.

Results: Preoperatively 10.0% of patients had hepatitis B or C, while 8.9% had cirrhosis. The indication for HR was benign (21.9%) or malignant (78.1%) disease. Among patients with a malignant indication, metastatic disease (63.7%) was more common than primary liver cancer (36.3%); a subset of patients (21.6%) had multiple tumors.  Preoperative treatment included neoadjuvant chemotherapy (23.8%), portal vein embolization (2.0%) and intra-arterial therapy (0.8%).  At surgery, most patients underwent an open HR (77.9%), while 21.2% and 0.9% underwent a laparoscopic or robotic procedure. The Pringle maneuver was used in 26.4% of patients.  While 6.1% of patients had a concomitant hepaticojejunostomy, 10.1% had a concurrent ablation performed with HR.  An operative drain was placed in half of patients (50.2%). Among the entire cohort, bile leak (6.6%) and liver insufficiency/failure (3.2%) were relatively uncommon.  A subset of patient (9.5%) did experience major liver-specific complications that required intervention (drainage of collection/abscess: 38.4%; stenting for biliary obstruction/leak: 21.2%; bilioma drainage: 18.4%).  

Conclusion: In addition to standard NSQIP variables, the new inclusion of liver-specific variables provides a unique opportunity to study national outcomes and practice patterns among patients undergoing HR. 

 

64.12 Severe acute pancreatitis in the community: confusion reigns

Posted on December 22, 2014

M. M. Dua1, D. Worhunsky1, R. Rumma1, T. Tran1, K. Hwa1, G. Poultsides1, J. Norton1, B. Visser1  1Stanford University School Of Medicine,Surgery,Stanford, CA, USA

Introduction: The management of acute pancreatitis (AP) has evolved through enhanced understanding of the pathophysiology and natural history of disease. Practice guidelines have emerged to address the issues of antibiotics (abx), nutrition, timing and/or need of intervention for local complications including pancreatic pseudocyst/necrosis. Despite these evidence-based recommendations, our hypothesis is that many hospitals still adhere to historical treatment practices despite the absence of clinical data.

Methods: Patients (pts) transferred to our institution with AP from 2010-14 were retrospectively studied. Pt charts, radiology reports, and procedure notes were reviewed to compare pre- vs post-transfer adherence to guidelines for the management of AP. Primary measures examined (that do not reflect current guidelines) included use of empiric abx, absence of enteral nutrition, drainage of pseudocysts, and intervention for pancreatic necrosis in the early phase (<4wks).

Results: Seventy-eight pts with AP were transferred to our institution from local community hospitals. Outside length of stay (LOS) was 9 (1-52) dys. Etiology of AP primarily included biliary 46%, alcoholic 17%, idiopathic 19%, and hypertriglyceridemia 11.5%; half of the cohort had a previous admission for AP. The admitting service was surgery (49%) or medicine (51%) with 17 pts admitted to the ICU. Pre-transfer, antibiotics (abx) were given to 51 pts; post-transfer, they were discontinued in 33 pts and started in 6 pts within 24hrs of admission (abx use pre vs post, 51 vs 24, p<0.001). Empiric abx for AP without evidence of infection was used in 36 pts pre- versus 9 pts (7 medicine; 2 surgery) post-transfer (p<0.001). Pts were NPO or on TPN in 89% of the cohort; this was reduced to 17% within 72hrs by starting oral diet or enteric feeds (enteral nutrition pre vs post, 9 vs 65 pts, p<0.001). Fifteen pts were transferred for a CT report of non-infected pseudocyst that "required" drainage; two pts had drains needing revision and 5 pts ultimately required intervention after transfer but were treated >4wks from initial episode of AP. Pre-transfer, 5 pts had debridement without evidence of infection (4 of these in early phase) which resulted in prolonged LOS (>21 dys) after transfer.

Conclusion: This study suggests that there is still great confusion in the treatment of AP in community hospitals. Primary principles in the care of these pts are not routinely followed despite established guidelines. Prophylactic abx are not recommended and enteral nutrition is encouraged to prevent infectious complications. Asymptomatic pseudocyst/necrosis do not warrant intervention; with superimposed infection, drainage or debridement should be delayed to >4wks. Increased dissemination of these guidelines is required to prevent lengthy hospitalizations and long term morbidity.

 

64.13 Outcome Of Abdominal And Colorectal Surgery In Patients With Left Ventricular Assistant Devices

Posted on December 22, 2014

A. A. Asban1, M. Traa1, J. Yoo1, N. Melnitchouk1  1Tufts Medical Center,Colorectal/Surgery/,Boston, MA, USA

Introduction:
Advances in left ventricular assistant device (LVAD) technology have contributed to significant improvements in patient quality of life and life expectancy. Assessment of the safety of abdominal and colorectal surgery in these patients is of particular importance due to their multiple comorbidities, hemodynamic status, and challenges with abdominal positioning of the LVAD driveline.

Methods:
We conducted a retrospective review of 13 patients with left ventricular assistant devices who had undergone abdominal (including colorectal) surgeries at Tufts Medical Center between January 2003 and December 2013. Data collected include type of device implanted, patients’ comorbidities, preoperative coagulation status, intraoperative complications, need for intraoperative blood products transfusion, type of surgery performed, postoperative complications and postoperative coagulation status. The 24-hour survival and thirty-day mortality rates were gathered in addition to their ultimate LVAD outcome.

Results:
A total of 13 patients underwent 17 operations during the study period. These included 5 colorectal (29.5%) and 12 other abdominal (70.5%) surgeries. The mean duration of LVAD support before surgery was 309 days. Thirteen (77%) patients had an LVAD as a bridge to transplant and 4 (24%) as destination therapy. A total of 8 (47%) surgeries were emergency and 9 (53%) elective surgeries. The most common intraoperative complication was arrhythmia (65% operations). The most common post-operative complications were bleeding in 3 (18%) operations, venous thromboembolism in 3 (18%) operations, and surgical site infection in 2 (12%) operations. The mean perioperative INR was 1.39. Intraoperative blood transfusion was required in 4 (24%) operations and FFP in 3 (18%) operations. Eight (44%) operations required blood transfusion in the first week post-operation. All patients survived the procedures with a 24-hour mortality rate of 12% and thirty-day mortality rate of 18%. 24-hour mortality was higher in emergency cases (25%) versus elective cases (0%). None of the deaths were from complications specific to the abdominal or colorectal surgery performed.

Conclusion
The study demonstrates the feasibility of abdominal and colorectal surgery in LVAD patients. Arrhythmia is the most common intraoperative complication while bleeding and venous thromboembolism are the most common postoperative complications. Emergency operations have higher 24-hour and 30 day mortality with most causes of death being nonspecific to the abdominal surgery performed. Surgeons should consider these expected good outcomes before resolving not to perform colorectal or other abdominal procedures on this higher-risk population.
 

64.15 Early Experience Quantifying Perfusion Time after Colectomy Using Endoscopic Fluorescence Imaging

Posted on December 22, 2014

C. Moore1, J. S. Turner1, C. E. Clark1  1Morehouse School Of Medicine,Department Of Surgery,Atlanta, GA, USA

Introduction

Anastomotic leak is seen in up to 20% of low colorectal anastomoses. Factors associated with higher risk of leak are poor blood flow to the anastomosis, contamination, anastomotic technique, tension, and declining distance from the anal verge. Here, we review our early experience with transanal, endoscopic fluorescence imaging to assess anastomotic perfusion time with correlation to early postoperative outcomes.

Methods

We prospectively collected data on adult patients undergoing elective colectomy or colostomy closure with an anastomosis to the rectum at a single teaching hospital. Demographics, operative, and postoperative data were retrospectively analyzed including time to maximal perfusion of the bowel proximal and distal to the anastomosis.  Perfusion was assessed using Pinpoint endoscopic fluorescence imaging system (Novadaq Technologies, Ontario, Canada).

Results

Eight consecutive laparoscopic and open colectomies with colorectal or ileorectal anastomosis were identified in a prospectively maintained database. Two board certified colorectal surgeons performed these cases with a resident at a single urban teaching hospital. The average age of the patients was 57 years old with a mean ASA score of 2.25 and BMI of 23.1. Complicated diverticulitis was the most common preoperative diagnosis followed by rectal and sigmoid adenocarcinoma. One patient had sigmoid colon cancer with attenuated polyposis necessitating a total colectomy. The distance between the anastomosis and the anal verge was on average 11.25 cm. There were no intraoperative complications or diversions. The mean perfusion time for colon or ileum after anastomosis was 50.375 seconds while the mean rectal perfusion time was 47.125 seconds. There were no significant differences between these two perfusion times.  One patient was found to have a leak postoperatively and returned to the operating room for colostomy formation and washout. The average length of stay was 9 days (+/- 6.4) with time to first BM of 5.4 days and first intake of enteral nutrition 5.1 days.

 

Conclusion

The use of transanal endoscopic fluorescence imaging is safe and should be considered in colorectal surgery. There are no differences in perfusion times between the proximal bowel and rectum following colorectal or ileorectal anastomosis. 

 

64.16 Emergency Upper GI Surgery in the United States: Burden of Disease and Ten-year trend – 2001-2010

Posted on December 22, 2014

J. S. Crystal1, V. Y. Dombrovskiy1, S. C. Gale1,2  1Robert Wood Johnson – Rutgers,Surgery,New Brunswick, NJ, USA 2East Texas Medical Center,Surgery,Tyler, TX, USA

Introduction:

Upper gastrointestinal (UGI) disorders, including peptic ulcer disease, are most commonly managed medically, yet patients presenting with emergent UGI conditions often require surgical evaluation and operative intervention. Using a large national database, we sought to assess the “burden of disease” for emergent presentation of UGI disorders and the need for surgical intervention during admission, over a ten-year period. 

Methods:
The Nationwide Inpatient Sample from 2001-2010 was queried for all patients admitted with emergent surgical illness (EGS) as recently defined by the AAST. The study population was selected using UGI-specific ICD-9 diagnosis and procedure codes and limited to patients admitted with urgent or emergent status. Operative rates, mortality, sepsis, and demographics were compiled. T-test and Cochran-Armitage trend test were used; p< 0.05 was significant.

Results:
During the 10-year study period 4,626,204 patients were admitted emergently to US hospitals with UGI disorders.  The average age was 64.7 years; 56% were female.  From 2000 to 2010, UGI admissions decreased as a percentage of total EGS (18.7% to 14.9%; p<0.0001) and while sepsis rates increased from 1.8% to 2.2% (p<0.0001), mortality declined from 2.6% to 1.6% (p<0.0001). During the study period, surgical rates increased markedly from 20.2% to 38.4% (p<0.0001) with a total of 1,438,649 patients (31.1%) requiring surgery. 

Conclusion:
While overall emergent admissions for UGI disorders have declined, surgical rates have increased significantly for those patients admitted. Although more patients present with sepsis, mortality has declined which may reflect improving critical care. These data suggest that general and acute care surgeons should address this growing emergent UGI disease burden with renewed emphasis on gastroduodenal procedures during surgical training and with future prospective inquiry.
 

64.17 Prospective Analysis of Mortality in Non-Trauma Patients Managed With Open Abdomen

Posted on December 22, 2014

S. Ahmad4, L. O’Meara3, E. Klyushnenkova2, T. M. Scalea2,3, J. Diaz2,3, B. R. Bruns2,3  2University Of Maryland School Of Medicine,Surgery,Baltimore, MD, USA 3R. Adams Cowley Shock Trauma Center,Surgery,Baltimore, MD, USA 4University Of Maryland,Department Of Surgery,Baltimore, MD, USA

Introduction:  Abbreviated laparotomy and open abdomen have shown acceptable mortality rates for non-trauma, acute care surgery (ACS) patients with abdominal catastrophe. However, these studies are retrospective and limited in size. In prospective fashion, we aimed to identify demographic and hospital course variables that were associated with in-hospital mortality. 

Methods:  All adult ACS patients managed with open abdomen from June 2013 until December 2013 were prospectively enrolled into an IRB-approved study at a single university medical center. Variables examined include demographics, Charlson Comorbidity Index (CCI), operative interventions, laboratory data, and outcomes. Wilcoxon rank sum, Fisher’s exact, and paired t tests were used to determine differences between survivors and non-survivors. 

Results: 58 patients (32 male, 26 female) were managed with open abdomen over a 6-month period with 25 in-hospital mortalities (43%). There was an association between age and mortality. Patients who died were older than those who survived (median 66 vs. 62 years, p=0.012). Mortality was not associated with gender, BMI (median 33.8), CCI, type of initial surgery, hospital or ICU length of stay (median 24.5 and 18 days, respectively), or time from initial surgery to death (median 16 days). The most common indication for laparotomy was mesenteric ischemia (n=12), followed closely by perforated viscus or pneumoperitoneum (n=11). No single indication for surgery was associated with increased mortality. The most common indication for open abdomen was need for damage control (n=30, 52%); however, presence of contamination or clinical indications for damage control (acidosis, coagulopathy, or hypothermia) had no association with mortality. Ventilator associated pneumonia was associated with increased mortality (p=0.003).  Lower preoperative platelet level (p=0.003) and hemoglobin level (p=0.046) were associated with increased mortality, but white blood cell count, INR, lactate, pH, and base deficit were not. 

Conclusion: This effort represents a large, single-institution, prospective cohort of ACS patients managed with open abdomen. Age, development of ventilator-associated pneumonia, platelet level, and to a lesser degree, hemoglobin level were associated with mortality. Contrary to previous studies, presence of intra-abdominal sepsis and damage control indicators at time of operation were not associated with increased mortality. ACS patients managed with open abdomen, in contrast to trauma, represent an older population with significant medical co-morbidities and a high likelihood of mortality.

 

64.18 Continuous topical irrigation improves delayed primary fascial closure of open septic abdomen

Posted on December 22, 2014

Q. TAO1, J. Ren2, B. Wang1, Y. Zheng1, J. Li2  1Affiliated Zhongda Hospital, Southeast University,General Surgery,Nanjing, JIANGSU, China 2Jinling Hospital, Nanjing University,General Surgery,Nanjing, JIANGSU, China

Introduction: Management of open abdomen (OA) is a formidable challenge, certainly when abdominal sepsis is present. Failure to achieve delayed primary fascial closure is one of the most common complications. This prospective cohort study aims to explore the influence of continuous topical irrigation on the management of open septic abdomen. 

Methods: The patients with abdominal sepsis who underwent OA using vacuum-assisted and mesh-mediated fascial traction (VAWCM) technique, were divided into the irrigation and control groups. The delayed primary fascial closure rate and other outcomes were compared between the two groups. 

Results:Between 2007 and 2013, 73 patients with open septic abdomen were treated with continuous topical irrigation and VAWCM, and 61 cases with only VAWCM. The overall delayed primary fascial closure rate in the irrigation group was significantly increased (63% vs 41%, p = 0.011). The mortality with OA was similar (24.6% vs 23%, p = 0.817). However, time to delayed primary fascial closure (p = 0.003) and length of stay in hospital (p = 0.022) in the survivals were significantly decreased in the irrigation group. In addition to topical irrigation (OR 1.453, 95% CI 1.222 – 4.927, p = 0.011), early enteral feeding (OR 3.357, 95% CI 1.479 – 7.619; p = 0.003) and restricted crystalloid fluid infusion (OR 2.611, 95% CI 1.296 – 5.261; p = 0.007) were independent influencing factors related to successful fascial closure of open septic abdomen.

Conclusion:Continuous topical irrigation could improve the delayed primary fascial closure, but could not decrease the mortality in the patients with open septic abdomen using VAWCM method. Early enteral feeding and restricted crystalloid fluid infusion might be recommended in such population.

 

64.19 Change in Functional Status and Quality of Life in Elders Admitted to an Acute Care Surgery Service

Posted on December 22, 2014

L. M. Warkentin1, T. F. Ali1, S. Gazala1, A. S. Wagg2, R. S. Padwal3, R. G. Khadaroo1  2University Of Alberta,Division Of Geriatric Medicine,Edmonton, ALBERTA, Canada 3University Of Alberta,Division Of General Internal Medicine,Edmonton, ALBERTA, Canada 1University Of Alberta,Department Of Surgery,Edmonton, ALBERTA, Canada

Introduction: Emergency surgery is increasingly common in older patients (≥ 65 years old).  Prior studies have shown that post-operative complications, in-hospital mortality, and an increased requirement for post-discharge care are more common in older patients. Patient-reported outcomes, including assessments of cognitive status, functional capacity and health-related quality of life (HRQL), are understudied. We conducted a six-month prospective cohort study in patients ≥ 65 years admitted to a specialized acute care and surgical emergency service at a single, publicly-funded, tertiary care center in Edmonton, Canada.

Methods: Participants, or their surrogates, completed the Abbreviated Mental Test Score-4 (AMTS-4) to measure cognitive status, Barthel Index to assess functional status, Vulnerable Elders Survey (VES-13) to examine risk of health deterioration, and EuroQol-5 Dimensional Scale (EQ-5D) for HRQL within 24 hours of admission to hospital and 6 months post-discharge. Paired t-tests or McNemar’s tests were used to assess differences in instrument scores between baseline and 6 months.

Results: One hundred fifty-five consecutive patients (including 16 surrogates) were enrolled. Sixteen (10%) patients died within 6 months of discharge and 116 (75%, including 18 surrogates) completed a follow-up assessment 6 months post-discharge. Fifty-two percent of patients were female and mean age was 77 years (SD 8). One hundred and two (66%) patients underwent surgery and 53 (34%) managed non-surgically. Cognitive status improved substantially over 6 months, 72 (52%) patients had AMTS scores showing cognitive impairment at baseline and 4 (4%) patients at 6 months (p < 0.001).  The mean (SD) Barthel Index score was 91.2 ± 14.1 at baseline and 92.8 ± 12.9 at 6 months (p > 0.05). Forty-seven (31%) patients had VES-13 scores indicating risk of health deterioration at baseline, and 33 (28%) patients at 6 months (p > 0.05). Mean (SD) EQ-Index score at baseline was 0.76 ± 0.24, and 0.86 ± 0.19 at 6 months (p > 0.05). Mean EQ-VAS score was 68.4 ± 20.3 at baseline and 72.0 ± 15.1 at 6 months (p > 0.05). Of the 116 patients that completed follow up, 17 patients (15%) required post-discharge escalation of care.

Conclusion: Elderly patients admitted to an acute care surgery service maintained functional capacity and improved quality of life 6-months post-discharge.  Initially low cognitive scores improved substantially in most patients, suggesting that the acute illness led to reversible cognitive impairment.  Our data suggest that the acute care surgery model leads to very low rates of cognitive impairment with good functional and HRQL outcomes in elderly patients presenting with emergent surgical illness.

64.20 Kidney-specific Morphomic Factors are Associated with AKI after Major General Surgery Operations

Posted on December 22, 2014

J. Li1, E. Chang1, N. C. Wang1, D. Cron1, P. Zhang1, S. C. Wang1  1University Of Michigan,Ann Arbor, MI, USA

Introduction: Acute kidney injury (AKI) occurs in approximately 20% of hospitalized adults in the US and is associated with increased morbidity, mortality, and costs for patients of general surgery procedures. Numerous studies have suggested the effectiveness of using patient CT imaging data to predict risk of AKI. Analytic Morphomics, a quantitative approach to processing cross-sectional imaging data, has shown utility in evaluating perioperative risk and postoperative outcome. Thus, we hypothesized that individualized morphomic factors would be significantly associated with AKI after major general surgery procedures.

Methods: 326 adult patients undergoing major, elective, intra-abdominal general surgery operations from January 2008 to September 2011 were selected. Patients undergoing isolated appendectomy or cholecystectomy procedures were excluded, as were patients with a preoperative creatinine value of 1.6 or greater, a history of ESRD, or a prior nephrectomy. Preoperative CT scans within 365 days prior to the operation were analyzed for kidney, vertebral, and fascia-associated parameters such as length, area, volume, and radiodensity. Demographic factors included patient age, gender, and BMI. Univariate analysis was utilized to determine the strength of the association between patient morphomic and demographic factors and development of postoperative AKI as defined by the KDIGO criteria within 30 days of the surgery.

Results: Of the 238 eligible patients, 31 (13%) developed AKI. Average pixel radiodensity of the total kidney (p<0.001) and kidney parenchyma (p<0.001) were found to be significantly lower for the group that developed postoperative AKI than for those that did not. No significant differences between AKI and non-AKI patients were found for total kidney volume (p=0.19), parenchymal volume (p=0.16), anterior body depth (p=0.32), and patient demographic factors (p>0.15). Table 1 contains the results of the univariate analysis including the means between the AKI and non-AKI group for several factors.

Conclusion: For patients undergoing noncardiac, nonvascular surgical procedures, our analysis showed that kidney pixel radiodensity factors are the most significant in predicting postoperative AKI risk.
 

65.01 Risks and Outcomes after Delirium following Cardiac Surgery: Analysis from a National Database

Posted on December 22, 2014

J. Idrees1, N. Schiltz1, E. E. Roselli1, A. Badjatiya1, D. Johnston1, E. G. Soltesz1  1Cleveland Clinic Foundation,Thoracic And Cardiovascular Surgery,Cleveland, OH, USA

Objective:  

Delirium commonly occurs following major surgery in the elderly. It is a known predictor of adverse outcome and increased hospitalization. In this study, we sought to determine the incidence of delirium, the prevalence of risk factors, and impact on healthcare resource use after cardiac surgery using a large administrative dataset.

Methods:

We identified 6,389,977 patients who underwent coronary bypass, valve, or thoracic aortic surgery between 1998 and 2011 from the Nationwide Inpatient Sample database. Multivariable regression was used to identify risk factors for post-operative delirium and risk-adjustment was performed in order to estimate the effect of a delirium on outcomes.

Results:

The incidence of postoperative delirium was 15.0 per 1000 cardiac surgery procedures. Patients with delirium were significantly older (mean age 72 vs. 66 years; p<0.001), male (69.7% vs. 67.6%; p<0.001), and had more Elixhauser comorbidities (6.2 vs. 4.0; p<0.001). Older age, Medicare or Medicaid coverage, and treatment at a teaching hospital were significantly associated with postoperative delirium, while hospital volume was not. Alcohol and drug abuse, depression, psychotic and neurologic disorder, anemia, fluid electrolyte imbalance and weight loss were independent clinical predictors of delirium. Patients who suffered delirium had a longer length of stay (13.8 days vs. 12.8; p<0.001), higher total costs ($52,843 vs. $50,586; p<.001), but lower risk of mortality (0.9% vs. 1.5%, p<0.001) than patients without delirium. Delirious patients were also more likely to be transferred to a skilled nursing facility or require home health care post discharge.

Conclusion:

Our results have shown that postoperative delirium in cardiac surgery patients is associated with increased resource use but lower mortality. Detailed pre-operative assessment and mitigation of certain risk factors can potentially improve outcomes and minimize resource utilization.

 

65.02 Comparing TAVI with Conventional AVR; First Experiences at the Liverpool Heart and Chest Hospital

Posted on December 22, 2014

R. E. Robinson1, T. Theologou1, M. Field1, R. Stables1, O. Al-Rawi1, M. Kuduvalli1, A. Oo1  1Liverpool Heart And Chest Hospital,Cardiothoracic Surgery,Liverpool, MERSEYSIDE, United Kingdom

Introduction:

Transcatheter aortic valve implantation (TAVI) is an alternative method for treatment of severe aortic valve stenosis in high-risk patients.

Recently the TAVI procedure has emerged as an effective alternative method for the treatment of severe aortic valve stenosis in high-risk patients. This innovative technique has been adopted in our institution since 2008 with a beneficial effect on patients who were excluded from conventional surgical management. The aim of this study is to compare the mortality and subsequent 4 year predicted survival rate of TAVI with traditional AVR.

Methods:
One hundred and sixty five patients were operated on between 2008 and 2013 at LHCH. The procedures were undertaken using both the transapical and transfemoral approaches. Retrospective observational cohort study analysis has been performed assessing the mortality and morbidity of this population. The outcome data from the TAVI patients was propensity matched to a conventional AVR group. A Kaplan Meier curve was developed allowing prediction of 4 year survival following the two procedures.  

Results:
122 TAVI patients were propensity matched with an AVR group. The TAVI group had a ICU length of stay of 1 day compared to 2 (p=0.001), an in hospital length of stay of 6 days compared to 8 (p= <0.001) in the AVR group. The TAVI group had an in-hospital mortality of 6 vs. 4 (p=0.52) and a 4-year mortality of 33 vs. 16 (p=0.007).

Conclusion:
TAVI patients have a shorter ICU and in hospital stay when compared to conventional AVR with no statistically significant difference in “in hospital mortality”. There is however a significant difference in 4 year survival when compared with AVR.

65.03 Simplifying Decisions With a New Risk Model for Cardiac Extracorporeal Membrane Oxygenation (ECMO)

Posted on December 22, 2014

G. Peigh1, H. T. Pitcher1, N. Cavarocchi1, H. Hirose1  1Thomas Jefferson University,Philadelphia, PA, USA

Introduction: While the use of cardiac ECMO is increasing in adult patients, an analysis of risk factors is still in its infancy.  Even though standard intensive care unit (ICU) risk scores such as Simplified Acute Physiology Score II (SAPS II), Sequential Organ Failure Assessment (SOFA) and Acute Physiology And Chronic Health Evaluation II (APACHE II), or disease specific scores such as Model for End-stage Liver Disease (MELD), Kidney Risk, Injury, Failure, Loss of Function, ESRD (RIFLE), Predicting death for Severe ARDS On VV ECMO (PRESERVE) and ECMOnet scores exist, they may not apply to cardiac ECMO patients as their risk factors differ from the variables these common scores use.

Methods: Between 2010 and 2014, 107 ECMO procedures were performed at our hospital. 73 procedures were to support patients on cardiac ECMO. Patient demographics, preoperative conditions, and survival were retrospectively analyzed with IRB approval.

Results: Cardiac ECMO was performed on 73 patients (47 males and 26 females) with a mean age of 48 ± 14 years.  The most common etiologies (#) for ECMO were acute myocardial infraction (19), acute on chronic heart failure (14), post-cardiotomy failure (13), malignant arrhythmia (11), and others (16).  The mean duration of ECMO support was 9.2 ± 6.1 days.  47/73 patients (64%) survived ECMO. 27/47 patients (57%) were discharged from the hospital.   The risk factors independently associated with death on ECMO were high lactate levels (p=0.02) and post-cardiotomy failure (p=0.03).  Pre-ECMO SAPS II, SOFA, APACHE II, MELD, RIFLE, ECMOnet and PRESERVE scores were not correlated with survival (Table). Analysis of pre-ECMO risk factors indicated that elevated lactate (>2.0 mmol/dl), metabolic acidosis (HCO3<20 mEq/L), renal dysfunction (RIFLE score of ‘injury’ or above), and having a post-cardiotomy failure predicted death. Applying this data into a new Simplified Cardiac ECMO Score (min 0, max 4) predicted survival (survivors 1.5 ± 1.1; non-survivors 2.4 ± 0.8; p=0.0006). The area under the curve (AUC) was 0.55 for SOFA, 0.60 for APACHE II, 0.54 for SAPS II, 0.57 for MELD, 0.49 for PRESERVE, 0.62 for ECMOnet, 0.60 for RIFLE, and 0.72 for the Simplified Cardiac ECMO score. Although our score has a higher AUC, a small sample size led to an underpowered study. Despite the fact that there is clinical significance, there was not statistical significance between the various AUC curves.

Conclusion: Common ICU or disease specific risk scores calculated for cardiac ECMO patients prior to the initiation of ECMO did not correlate with ECMO survival. Although the Simplified Cardiac ECMO Score needs to be further investigated, it helps predict futile efforts in high-risk populations.

 

65.05 Thoracoscopic Ablation of Persistent Atrial Fibrillation on the Beating Heart

Posted on December 22, 2014

A. Muehle1, D. Chou1, M. D. Te Winkel1, A. Khoynezhad1  1Cedars-Sinai Medical Center,Cardiothoracic Surgery,Los Angeles, CA, USA

Introduction:
Atrial fibrillation (AF) is the most common sustained arrhythmia and is associated with a nearly five-fold increased risk for stroke as well as over two-fold increased risk of death. For symptomatic drug-refractory AF, percutaneous ablation has been used with good success in paroxysmal AF. For patients with persistent AF, the results of catheter ablation are poor. Therefore, surgical AF using minimal-invasive approaches is offered to this cohort. We analyzed our data in thoracoscopically-performed ablation of AF.

Methods:
From January 2012 through March 2014, a total of 27 symptomatic and drug-refractory patients with lone AF underwent a thoracoscopic epicardial ablation on the beating heart using the bipolar radiofrequency energy source. All of them had persistent long-standing AF and besides one patients also a history of at least one percutaneous catheter ablation. Epicardial ablation was performed on 19 men (70%) and 8 women (30%), with a mean age of 64 (range 47 to 82) years. After the ablation, in all patients entrance and exit block was confirmed, the ganglionic plexi were ablated, and the left atrial appendage was excluded.

Results:
The mean operative time was 4.6 h and the mean postoperative length of stay was 6.5 days. In one patient cardiopulmonary bypass was used due to a laceration of the left atrial appendage. There were no hospital deaths, strokes or myocardial infarctions. The heart rhythm was documented in all of them besides 7 patients with a holder-ECG with duration of at least of 7-days. In 2 patients the postoperatively persistent AF was treated with radiofrequency ablation. Pacemaker implantation was done in 3 patients (11%) due to bradycardia (n=2) and sick-sinus syndrome (n=1). There were 2 late deaths (7%) one due to pulmonary embolism (PE) with a previous history of PE and the other patient died due to a major stroke. The follow-up was completed in 81.5% with a mean length of 11 months (range 90 to 793 days) with a freedom of AF in 85%.

Conclusion:
Thoracoscopic AF ablation on the beating heart for treatment of AF is technical feasible and achieve high success rates with low procedure-related morbidity in early follow-up.
 

63.11 Alum Pretreatment Enhances Protective Immunity and Improves Survival in Neonatal Polymicrobial Sepsis

Posted on December 22, 2014

S. D. Larson1, A. L. Cuenca1, B. E. Szpila2, B. Mathias1, A. G. Cuenca2, L. F. Gentile2, P. A. Efron2, L. L. Moldawer2  1University Of Florida,Pediatric Surgery, Department Of Surgery, UF College Of Medicine,Gainesville, FL, USA 2University Of Florida,Department Of Surgery, UF College Of Medicine,Gainesville, FL, USA

Introduction: Severe infection and sepsis during the neonatal period is a global health care issue with over 1 million deaths annually. Contributing to this high mortality is the neonate’s functionally distinct innate immune response. In previous studies, we have demonstrated that adjuvant activation via the TLR4 receptor improves outcomes in neonatal sepsis. Unfortunately, TLR agonists (e.g. LPS, resiquimod) are contraindicated in clinical use due to adverse side-effects. Aluminum (alum) salts are currently used as adjuvants in pediatric vaccines to improve immune responses; however, the mechanisms by which alum mediate these responses are incompletely understood. Therefore, the purpose of our study was two-fold: 1) As the neonate is more dependent on innate immunity, we examined whether pretreating neonates with alum wound improve survival and 2) to determine if alum enhances the myelopoietic response to polymicrobial sepsis.

Methods: 5-7 day old (neonate) C57BL/6J (B6) mice received intraperitoneal (IP) administration of cecal slurry (CS; LD25-45) to induce intra-abdominal polymicrobial sepsis. Neonates received either no pretreatment (control) or 20 µg aluminum hydroxide (alum) via IP injection 24 h prior to CS administration. Following injection of CS, mice were observed for 7 days to determine survival. Bone marrow and splenocytes were harvested at baseline (0 h) and 36 h following CS. Peritoneal washes and blood samples were collected at 0, 2, 6 and 24 h following sepsis. Harvested cells were analyzed by flow cytometry for phenotype.

Results: Neonates pretreated with alum had significantly improved survival compared to groups receiving CS alone (p<0.001). Alum pretreated mice had significant expansion of hematopoietic stem cells (Lin–ska+c-kit+ or LSKs) in the bone marrow and spleen 36 h following sepsis compared to controls (1.98±0.25 vs. 0.61±0.05 (p<0.0001), 1.59±0.51 vs. 0.42±0.04 (p<0.001), respectively). At 2 h following sepsis, pretreated neonates had increased populations of macrophages and natural killer (NK) cells in the peritoneum compared to controls (7.57±0.81% vs. 1.86±0.92% (p=0.001), 11.6±2.97% vs. 0.94±0.46% (p<0.003), respectively). Total percentage of peritoneal NK cells remained increased at 6 h following sepsis in pretreated neonates (4.97±2.37% vs. 0.66±0.27%; p=0.03).

Conclusion: We demonstrate here for the first time that neonates pretreated with alum have improved survival following polymicrobial sepsis. Alum pretreatment leads to a dramatic increase in bone marrow and splenic LSKs, peritoneal macrophages and NK cells, likely influencing this survival advantage. Finally, alum pretreatment appears to play a previously undescribed role in the neonatal myelopoietic response to sepsis.

 

63.13 Pilot Survey to Assess the Burden of Surgical Conditions in a Peri–Urban District in Uganda

Posted on December 22, 2014

E. K. Butler1, T. Tran2, A. Fuller2,3, F. Makumbi5, S. Luboga7, S. Kisakye5, M. Haglund2,9, J. Chipman10, M. Galukande11  1University Of Minnesota,Medical School,Minneapolis, MN, USA 2Duke University Medical Center,Global Health Institute,Durham, NC, USA 3Duke University Medical Center,Medical School,Durham, NC, USA 5Makerere University,School Of Public Health,Kampala, , Uganda 7Makerere University,Department Of Anatomy,Kampala, , Uganda 9Duke University Medical Center,Division Of Neurosurgery,Durham, NC, USA 10University Of Minnesota,Department Of Surgery,Minneapolis, MN, USA 11Makerere University,Department Of Surgery,Kampala, , Uganda

Introduction:  Low- and middle-income countries are challenged by the dual burden of infectious and non-communicable diseases, including those requiring surgical care. Globally, it is estimated that 11% of all disability adjusted life-years lost result from conditions requiring surgery, however little is known about country or disease specific burden. We piloted a household-based survey to assess the burden of surgical conditions in a large peri-urban district of Uganda to estimate the population prevalence of surgical conditions and to identify logistical challenges of such a survey.

Methods:  A total of 55 households in 5 randomly selected enumeration areas in the peri-urban district of Wakiso, Uganda were systematically sampled to complete the Surgeons OverSeas Assessment of Surgical Need survey. The head of household completed demographic and recent household death information, and 2 randomly selected individuals in each household completed a head-to-toe questionnaire on possible surgical conditions.  The current and lifetime prevalence of surgical conditions and the proportion of recent household deaths that could be attributed to surgery were determined.

Results: Eight of 96 participants (8.3%) had an existing surgical condition, 6 of whom were currently in need of surgical care.  The lifetime prevalence of surgical conditions was 28% (27/96) and 26% (7/27) of those individuals had some degree of disability due to their condition. The most common barrier to access to care was lack of financial resources. Two of 3 household deaths were surgically associated. The average time required to complete each household survey was 36 minutes. The main challenges were efficient coordination with local team members and government officials for completion of enumeration areas and language barriers. 

Conclusion: This pilot study in a peri-urban district in Uganda showed a similar prevalence of surgical conditions compared to previous implementations of this survey in Rwanda and Sierra Leone. There were a limited number of addressable challenges in implementation of the pilot. A complete, nationwide study to assess the burden of surgical conditions is both feasible and necessary to further characterize the met and unmet need for surgical care in Uganda. 

63.14 Free Care Is Not Enough: Patient Navigation Increases Access to Surgical Care in Rural Haiti

Posted on December 22, 2014

A. C. Matousek1,3, J. Denike2, S. R. Addington1, C. Exe2, R. R. Jean-Louis2, J. G. Meara3, R. Riviello1,3  1Brigham And Women’s Hospital,The Center For Surgery And Public Health,Boston, MA, USA 2Hospital Albert Schweitzer,Deschapelles, ARTIBONITE, Haiti 3Harvard Medical School,Program In Global Surgery And Social Change, Department Of Global Health And Social Medicine,Boston, MA, USA

Introduction:  In the catchment area of an NGO hospital in rural Haiti, the surgical service rate among the poorer population from mountainous areas is much lower than in the relatively wealthier population living in valley areas, despite the presence of a free care program for the poor.  Potential additional barriers for mountain patients may include low literacy levels, lack of awareness of the free care program, unfamiliarity with the hospital system, long distances to travel and high opportunity costs. We sought to increase the elective operation rate among patients from a mountain population with a surgically treatable condition using a patient navigation (PN) program.

Methods:  A surgical resident screened potential subjects with a physical examination at a mobile health post in the mountain region. Subjects responded to a questionnaire regarding their symptoms and knowledge of the free care program. Subjects with more severe disease were given priority. Patient navigators (PNs) were trained to guide subjects through an initial clinic visit for elective surgery from March 10 to July 1, 2014. A control period from December 10 to March 1, 2014 was used for comparison. The rate of elective operations per 10,000 population was measured across a mountain control group, a valley control group and the PN group both during the control period and during the PN program.

Results: Surgical conditions were found in 86 of 136 subjects who presented for screening (63.7%). Subjects had experienced symptoms for more than two years in 70.3% of cases. Less than 5% of subjects claimed any prior knowledge of the free care program and only 34% could write their name. PNs successfully guided 41 high priority subjects who received operations during the study period. After controlling for temporal trends across the two control groups, the elective operation rate in the PN group increased from 0.91 operations per 10,000 population in the control period to 5.05 operations per 10,000 population with PN (p = 0.002). The rate of urgent surgical operations remained the same across all three groups.

Conclusions: A PN program was able to increase the rate of elective surgical operations for a vulnerable population in rural Haiti. While additional barriers remain, PN appears to be an effective augmentation to free care programs to ensure equitable access to surgical care for the poor.

 

63.15 The Struggle for Equity: An Examination of Surgical Services at Two NGO Hospitals in Rural Haiti

Posted on December 22, 2014

A. C. Matousek1,4, S. R. Addington1, R. R. Jean Louis2, J. Hamiltong Pierre3, J. Fils4, M. Hoyler4, S. B. Matousek1,5, J. Pyda4, P. Farmer4, J. G. Meara4, R. Riviello1,4  1Brigham And Women’s Hospital,The Center For Surgery And Public Health,Boston, MA, USA 2Hopital Albert Schweitzer,Surgery,Deschapelles, ArtIBONITE, Haiti 3Hopital Bon Sauveur,Zanmi Lasante,Cange, CENTRAL, Haiti 4Harvard Medical School,Program in Global Surgery And Social Change, Department Of Global Health And Social Medicine,Boston, MA, USA 5Boston University School Of Public Health,Department Of Health Policy And Management,Boston, MA, USA

Introduction:  Health systems must deliver care equitably in order to serve the poor.  Both Hospital A and Hospital B have longstanding commitments to provide equitable surgical care in rural Haiti. Hospital A charges fees that demonstrate a preference for the rural population near the hospital, with free care available for the poorest. Hospital B does not charge fees. The two hospitals are otherwise similar in surgical capacity and rural location. 




Methods:  We retrospectively reviewed operative case-logs at both hospitals from June 1 to Aug 31, 2012. The records were compared by total number of operations, geographic distribution of patients and number of elective operations. Hospital A defines a rural service area comprised of the nearby administrative units.  Hospital B does not define a service area. To align with its mission to provide a preferential option for the poor, we defined the service area of Hospital B to include the entire nation except urban areas that contain other hospitals providing surgical care. For Hospital A, we analyzed the number of operations performed on patients from the most and least poor regions within the service area.

Results: Hospital A performed 348 operations and Hospital B performed 410 operations during the study period.  Rural patients received 86% of operations at Hospital A compared to 53% at Hospital B (p<0.0001). Urban patients with elective conditions accounted for only 1% of all operations at Hospital A compared to 15% at Hospital B (p<0.0001).   Within its rural service area, Hospital A performed 10.1 operations per 10,000 residents of less impoverished locations compared to 4.0 operations per 10,000 residents of severely destitute areas (p<0.0001).

Conclusion: Using fees as part of an equity strategy will likely disadvantage the poorest patients, while providing care without fees may encourage patients to travel from urban areas that contain other hospitals. Health systems striving to serve the poor should continually evaluate and seek to improve equity, even within systems that provide free care.

 

63.16 The Impact Factor of Social Media: Lessons from The Lancet Commission on Global Surgery

Posted on December 22, 2014

J. N. Riesel1,2,6, J. S. Ng-Kamstra2,3,6, S. L. Greenberg2,4,6, N. P. Raykar2,5,6, J. G. Meara2,6  1Massachusetts General Hospital,Department Of Surgery,Boston, MA, USA 2Children’s Hospital Boston,Department Of Plastic And Oral Surgery,Boston, MA, USA 3University of Toronto,Department Of Surgery,Toronto, Ontario, Canada 4Medical College Of Wisconsin,Department Of Surgery,Milwaukee, WI, USA 5Beth Israel Deaconess Medical Center,Department Of Surgery,Boston, MA, USA 6Harvard School Of Medicine,Program In Global Surgery And Social Change,Brookline, MA, USA

Introduction:
The Lancet Commission on Global Surgery (LCoGS) addresses the critical need for surgical care worldwide in an era marked by the pervasive use of social media (SM). To rally the global health community, engage key stakeholders, and strengthen the content and impact of the Commission findings, LCoGS launched a SM campaign capitalizing on the widespread use of SM by both academics and the general public. LCoGS Twitter and Facebook accounts were launched in December 2013.  Live updates from commission meetings, current articles, news, and advocacy pertaining to global surgery were posted. However, at the conception of this campaign, little was known about strategies for optimizing the reach and engagement of global surgery SM posts.

Methods:
A retrospective review of all original LCoGS Twitter and Facebook posts was performed. Posts were characterized across 25 suspected explanatory content and timing variables. The impact of each Facebook post was measured as engagement and reach, where engagement was measured as the number of unique users who clicked within the post, and reach was measured as the number of unique users who received the post within their news feed. For Twitter analysis, public engagement was measured as total number of interactions (Likes, Retweets, and Favorites) with each tweet. Univariate and multivariate regression analyses were used to determine the most impactful content in LCoGS’ SM posts.

Results:
391 Twitter and 113 Facebook posts from a 7-month period were analyzed. For Facebook, time of day of post, use of the “#GlobalSurgery” hashtag, and presence of a hyperlink were significantly correlated with SM reach. Messages posted after 4PM had more than twice the reach of those posted before 8AM. For Twitter, posts including links, hashtags, statistics, and expressions of opinion significantly correlated with increased engagement. Tweets containing specific global surgery statistics had an average of 2.49 times the number of interactions than those without. For both platforms, univariate analysis revealed trends of increased engagement when women’s issues, children’s issues, or photographic images were included in the post.

Conclusion:
SM stimulates a global, multidirectional convergence that can enhance engagement, advocacy, and academic pursuits.  Thoughtful construction of SM posts can lead to increased engagement with desired stakeholders. Both content and timing influence the reach of SM posts for a given audience and should be considered when constructing a social media campaign.  Although we can investigate the reach and engagement stimulated by the LCoGS SM posts, we cannot determine the effect the posts have on human behavior or scientific pursuits.  Still, as the groundswell of dependency on SM as a source of current news, public opinion, and scientific work continues to grow, SM can be considered a cardinal tool to both the general public and the surgeon scientist alike.
 

63.17 Informed Consent for Surgical Missions in the Developing World: The Patient Perspective

Posted on December 22, 2014

C. D. Sutton1, J. D. Sharma2, G. C. Lynde1  1Emory University School Of Medicine,Anesthesiology,Atlanta, GA, USA 2Emory University School Of Medicine,Surgery,Atlanta, GA, USA

Introduction:

The continuum between paternalism and autonomy has garnered increasing attention in the recent past and is the subject of debate in current medical ethics literature. Much of the discussion is centered on informed consent and the evolving Western values of the last century. It is clear that the medical community believes that universal standards for bioethics including the tenant of informed consent should be upheld in the setting of humanitarian medical missions. However, differences in culture, language, and infrastructure complicate the issue of how to best approach the consent process for short-term surgical missions.

The obstacles to obtaining informed consent in the global health setting have been thoroughly discussed in the literature, but few if any studies have investigated these issues from the patient’s perspective. Our study sought to understand the patient’s experience of giving consent in the setting of a two-week surgical mission to Haiti. Using a survey, we addressed questions regarding interaction, communication, and our consent process in an effort to find a more culturally and ethically appropriate method of obtaining consent.

Methods:  A survey was created and administered using the iPad app Quicktap Survey. The survey was written in English at a Flesch Kincaid Grade Level of 2.2 and was then translated into Haitian Creole. An interpreter was present for every survey to help patients read, understand, and answer the questions. All patients selected for surgery were offered the opportunity to participate after the consent process for their surgery was complete. Data was collected during both weeks of the two-week trip.

Results:

We surveyed 55 patients, of whom 80% were male and the average age was 40.9. 72% of patients had completed primary school or less, and 75% had never had surgery before.

Regarding communication, 93% reported that they felt comfortable asking questions of an American doctor, and 67% felt equally or more comfortable asking questions to an American doctor as compared to a Haitian doctor. However, only 47% reported finding communication using an interpreter to be easy, and 9% felt that their questions were not always understood when using an interpreter.

While 82% reported knowing the risks of their surgery, when asked to identify these risks 27% selected one or more sham answers. Regarding the idea of giving consent, 98% of patients felt it was important to understand their surgery and its risks, and 55 of 55 patients stated that signing a consent form that showed this understanding was important.

Conclusion: Our results highlight several areas needing improvement in the consent process including clearer risk communication and more effective use of interpreters. Our survey demonstrates that the process of obtaining informed consent for surgery is important to patients in the setting of short-term medical missions and suggests that continued efforts to improve our methods are critical.

« Previous Page
Next Page »