Predicting Colonic Neoplasia Surgical Complications: A Machine Learning Approach.

BACKGROUND: A range of statistical approaches have been used to help predict outcomes associated with colectomy. The multifactorial nature of complications suggests that machine learning algorithms may be more accurate in determining postoperative outcomes by detecting nonlinear associations, which are not readily measured by traditional statistics. OBJECTIVE: The aim of this study was to investigate the utility of machine learning algorithms to predict complications in patients undergoing colectomy for colonic neoplasia. DESIGN: Retrospective analysis using decision tree, random forest, and artificial neural network classifiers to predict postoperative outcomes. SETTINGS: National Inpatient Sample database (2003-2017). PATIENTS: Adult patients who underwent elective colectomy with anastomosis for neoplasia. MAIN OUTCOME MEASURES: Performance was quantified using sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve to predict the incidence of anastomotic leak, prolonged length of stay, and inpatient mortality. RESULTS: A total of 14,935 patients (4731 laparoscopic, 10,204 open) were included. They had an average age of 67 ± 12.2 years, and 53% of patients were women. The 3 machine learning models successfully identified patients who developed the measured complications. Although differences between model performances were largely insignificant, the neural network scored highest for most outcomes: predicting anastomotic leak, area under the receiver operating characteristic curve 0.88/0.93 (open/laparoscopic, 95% CI, 0.73-0.92/0.80-0.96); prolonged length of stay, area under the receiver operating characteristic curve 0.84/0.88 (open/laparoscopic, 95% CI, 0.82-0.85/0.85-0.91); and inpatient mortality, area under the receiver operating characteristic curve 0.90/0.92 (open/laparoscopic, 95% CI, 0.85-0.96/0.86-0.98). LIMITATIONS: The patients from the National Inpatient Sample database may not be an accurate sample of the population of all patients undergoing colectomy for colonic neoplasia and does not account for specific institutional and patient factors. CONCLUSIONS: Machine learning predicted postoperative complications in patients with colonic neoplasia undergoing colectomy with good performance. Although validation using external data and optimization of data quality will be required, these machine learning tools show great promise in assisting surgeons with risk-stratification of perioperative care to improve postoperative outcomes. See Video Abstract . PREDICCIN DE LAS COMPLICACIONES QUIRRGICAS DE LA NEOPLASIA DE COLON UN ENFOQUE DE MODELO DE APRENDIZAJE AUTOMTICO: ANTECEDENTES:Se han utilizado una variedad de enfoques estadísticos para ayudar a predecir los resultados asociados con la colectomía. La naturaleza multifactorial de las complicaciones sugiere que los algoritmos de aprendizaje automático pueden ser más precisos en determinar los resultados posoperatorios al detectar asociaciones no lineales, que generalmente no se miden en las estadísticas tradicionales.OBJETIVO:El objetivo de este estudio fue investigar la utilidad de los algoritmos de aprendizaje automático para predecir complicaciones en pacientes sometidos a colectomía por neoplasia de colon.DISEÑO:Análisis retrospectivo utilizando clasificadores de árboles de decisión, bosques aleatorios y redes neuronales artificiales para predecir los resultados posoperatorios.AJUSTE:Base de datos de la Muestra Nacional de Pacientes Hospitalizados (2003-2017).PACIENTES:Pacientes adultos sometidos a colectomía electiva con anastomosis por neoplasia.INTERVENCIONES:N/A.PRINCIPALES MEDIDAS DE RESULTADO:El rendimiento se cuantificó utilizando la sensibilidad, especificidad, precisión y la característica operativa del receptor del área bajo la curva para predecir la incidencia de fuga anastomótica, duración prolongada de la estancia hospitalaria y mortalidad de los pacientes hospitalizados.RESULTADOS:Se incluyeron un total de 14.935 pacientes (4.731 laparoscópicos, 10.204 abiertos). Presentaron una edad promedio de 67 ± 12,2 años y el 53% eran mujeres. Los tres modelos de aprendizaje automático identificaron con éxito a los pacientes que desarrollaron las complicaciones medidas. Aunque las diferencias entre el rendimiento del modelo fueron en gran medida insignificantes, la red neuronal obtuvo la puntuación más alta para la mayoría de los resultados: predicción de fuga anastomótica, característica operativa del receptor del área bajo la curva 0,88/0,93 (abierta/laparoscópica, IC del 95%: 0,73-0,92/0,80-0,96); duración prolongada de la estancia hospitalaria, característica operativa del receptor del área bajo la curva 0,84/0,88 (abierta/laparoscópica, IC del 95%: 0,82-0,85/0,85-0,91); y mortalidad de pacientes hospitalizados, característica operativa del receptor del área bajo la curva 0,90/0,92 (abierto/laparoscópico, IC del 95%: 0,85-0,96/0,86-0,98).LIMITACIONES:Los pacientes de la base de datos de la Muestra Nacional de Pacientes Hospitalizados pueden no ser una muestra precisa de la población de todos los pacientes sometidos a colectomía por neoplasia de colon y no tienen en cuenta factores institucionales y específicos del paciente.CONCLUSIONES:El aprendizaje automático predijo con buen rendimiento las complicaciones postoperatorias en pacientes con neoplasia de colon sometidos a colectomía. Aunque será necesaria la validación mediante datos externos y la optimización de la calidad de los datos, estas herramientas de aprendizaje automático son muy prometedoras para ayudar a los cirujanos con la estratificación de riesgos de la atención perioperatoria para mejorar los resultados posoperatorios. (Traducción-Dr. Fidel Ruiz Healy ).

Feasibility and comparison of laparoscopic laser speckle contrast imaging to near-infrared display of indocyanine green in intraoperative tissue blood flow/tissue perfusion in preclinical porcine models.

OBJECTIVE: To determine if laser speckle contrast imaging (LSCI) mitigates variations and subjectivity in the use and interpretation of indocyanine green (ICG) fluorescence in the current visualization paradigm of real-time intraoperative tissue blood flow/perfusion in clinically relevant scenarios. METHODS: De novo laparoscopic imaging form-factor detecting real-time blood flow using LSCI and blood volume by near-infrared fluorescence (NIRF) of ICG was compared to ICG NIRF alone, for dye-less real-time visualization of tissue blood flow/perfusion. Experienced surgeons examined LSCI and ICG in segmentally devascularized intestine, partial gastrectomy, and the renal hilum across six porcine models. Precision and accuracy of identifying demarcating lines of ischemia/perfusion in tissues were determined in blinded subjects with varying levels of surgical experience. RESULTS: Unlike ICG, LSCI perfusion detection was real time (latency < 150 ms: p < 0.01), repeatable and on-demand without fluorophore injection. Operating surgeons (n = 6) precisely and accurately identified concordant demarcating lines in white light, LSCI, and ICG modes immediately. Blinded subjects (n = 21) demonstrated similar spatial-temporal precision and accuracy with all three modes ≤ 2 min after ICG injection, and discordance in ICG mode at ≥ 5 min in devascularized small intestine (p < 0.0001) and in partial gastrectomy (p < 0.0001). CONCLUSIONS: Combining LSCI for near real-time blood flow detection with ICG fluorescence for blood volume detection significantly improves precision and accuracy of perfusion detection in tissue locations over time, in real time, and repeatably on-demand than ICG alone.

Utility and usability of laser speckle contrast imaging (LSCI) for displaying real-time tissue perfusion/blood flow in robot-assisted surgery (RAS): comparison to indocyanine green (ICG) and use in laparoscopic surgery.

BACKGROUND: Utility and usability of laser speckle contrast imaging (LSCI) in detecting real-time tissue perfusion in robot-assisted surgery (RAS) and laparoscopic surgery are not known. LSCI displays a color heatmap of real-time tissue blood flow by capturing the interference of coherent laser light on red blood cells. LSCI has advantages in perfusion visualization over indocyanine green imaging (ICG) including repeat use on demand, no need for dye, and no latency between injection and display. Herein, we report the first-in-human clinical comparison of a novel device combining proprietary LSCI processing and ICG for real-time perfusion assessment during RAS and laparoscopic surgeries. METHODS: ActivSight™ imaging module is integrated between a standard laparoscopic camera and scope, capable of detecting tissue blood flow via LSCI and ICG in laparoscopic surgery. From November 2020 to July 2021, we studied its use during elective robotic-assisted and laparoscopic cholecystectomies, colorectal, and bariatric surgeries (NCT# 04633512). For RAS, an ancillary laparoscope with ActivSight imaging module was used for LSCI/ICG visualization. We determined safety, usability, and utility of LSCI in RAS vs. laparoscopic surgery using end-user/surgeon human factor testing (Likert scale 1-5) and compared results with two-tailed t tests. RESULTS: 67 patients were included in the study-40 (60%) RAS vs. 27 (40%) laparoscopic surgeries. Patient demographics were similar in both groups. No adverse events to patients and surgeons were observed in both laparoscopic and RAS groups. Use of an ancillary laparoscopic system for LSCI/ICG visualization had minimal impact on usability in RAS as evidenced by surgeon ratings of device usability (set-up 4.2/5 and form-factor 3.8/5). LSCI ability to detect perfusion (97.5% in RAS vs 100% in laparoscopic cases) was comparable in both RAS and laparoscopic cases. CONCLUSIONS: LSCI demonstrates comparable utility and usability in detecting real-time tissue perfusion/blood flow in RAS and laparoscopic surgery.

Real-time quantification of intestinal perfusion and arterial versus venous occlusion using laser speckle contrast imaging in porcine model.

PURPOSE: Real-time intraoperative perfusion assessment may reduce anastomotic leaks. Laser speckle contrast imaging (LSCI) provides dye-free visualization of perfusion by capturing coherent laser light scatter from red blood cells and displays perfusion as a colormap. Herein, we report a novel method to precisely quantify intestinal perfusion using LSCI. METHODS: ActivSight™ is an FDA-cleared multi-modal visualization system that can detect and display perfusion via both indocyanine green imaging (ICG) and LSCI in minimally invasive surgery. An experimental prototype LSCI perfusion quantification algorithm was evaluated in porcine models. Porcine small bowel was selectively devascularized to create regions of perfused/watershed/ischemic bowel, and progressive aortic inflow/portal vein outflow clamping was performed to study arterial vs. venous ischemia. Continuous arterial pressure was monitored via femoral line. RESULTS: LSCI perfusion colormaps and quantification distinguished between perfused, watershed, and ischemic bowel in all vascular control settings: no vascular occlusion (p < 0.001), aortic occlusion (p < 0.001), and portal venous occlusion (p < 0.001). LSCI quantification demonstrated similar levels of ischemia induced both by states of arterial inflow and venous outflow occlusion. LSCI-quantified perfusion values correlated positively with higher mean arterial pressure and with increasing distance from ischemic bowel. CONCLUSION: LSCI relative perfusion quantification may provide more objective real-time assessment of intestinal perfusion compared to conventional naked eye assessment by quantifying currently subjective gradients of bowel ischemia and identifying both arterial/venous etiologies of ischemia.

Real-time quantification of bowel perfusion using Laparoscopic Laser Speckle Contrast Imaging (LSCI) in a porcine model.

BACKGROUND/PURPOSE: Real-time quantification of tissue perfusion can improve intraoperative surgical decision making. Here we demonstrate the utility of Laser Speckle Contrast Imaging as an intra-operative tool that quantifies real-time regional differences in intestinal perfusion and distinguishes ischemic changes resulting from arterial/venous obstruction. METHODS: Porcine models (n = 3) consisted of selectively devascularized small bowel loops that were used to measure the perfusion responses under conditions of control/no vascular occlusion, arterial inflow occlusion, and venous outflow occlusion using laser speckle imaging and indocyanine green fluoroscopy. Laser Speckle was also used to assess perfusion differences between small bowel antimesenteric-antimesenteric and mesenteric-mesenteric anastomoses. Perfusion quantification was measured in relative perfusion units calculated from the laser speckle perfusion heatmap. RESULTS: Laser Speckle distinguished between visually identified perfused, watershed, and ischemic intestinal segments with both color heatmap and quantification (p < .00001). It detected a continuous gradient of relative intestinal perfusion as a function of distance from the stapled ischemic bowel edge. Strong positive linear correlation between relative perfusion units and changes in mean arterial pressure resulting from both arterial (R2 = .96/.79) and venous pressure changes (R2 = .86/.96) was observed. Furthermore, Laser Speckle showed that the antimesenteric anastomosis had a higher perfusion than mesenteric anastomosis (p < 0.01). CONCLUSIONS: Laser Speckle Contrast Imaging provides objective, quantifiable tissue perfusion information in both color heatmap and relative numerical units. Laser Speckle can detect spatial/temporal differences in perfusion between antimesenteric and mesenteric borders of a bowel segment and precisely detect perfusion changes induced by progressive arterial/venous occlusions in real-time.

Health disparities attributed to Medicare-Medicaid dual-eligible status in patients with peripheral arterial disease.

OBJECTIVE: Peripheral arterial disease (PAD) is a prevalent and debilitating disease that can be effectively treated by surgical revascularization. However, Medicare-Medicaid dual-eligible patients have experienced worse long-term outcomes, notably higher rates of amputation and mortality, relative to other insurance groups. In the present study, we investigated how insurance status can perpetuate health disparities in PAD outcomes. METHODS: The National Inpatient Sample was queried from 2000 to 2011 for patients aged ≥18 years with PAD who had undergone surgical revascularization with hospitalization. Patients were stratified by insurance status, and dual-eligible patients were compared with Medicare-only, Medicaid-only, private insurance, and self-pay patients. Multivariable regression analysis was performed to assess the effect of dual-eligible status on postoperative outcomes such as inpatient mortality, complications, and favorable discharge (home or home with services). RESULTS: A total of 771,790 hospitalizations were included in the present analysis and stratified by insurance type. Dual-eligible patients had the highest rates of major (32%) and extreme (11%) severity of illness and the highest rates of major (19%) and extreme (6%) risk of mortality among all insurance groups (P < .001). Dual-eligibility status was independently associated with reduced odds of favorable discharge relative to all patients (P < .001) and increased length of stay relative to Medicare-only (P = .002) and private-payor groups (P < .001). Although dual-eligible patients had increased mortality odds relative to the Medicaid-only and self-pay groups, they did not have significantly different odds of perioperative complications relative to all other insurance groups. CONCLUSIONS: Medicare-Medicaid dual-eligible patients with PAD had had more severe clinical presentations, a greater risk of extended hospitalizations, and a lower likelihood of discharge to home, relative to patients without dual eligibility. Further studies are needed to examine the link between discharge disposition and disparities in healthcare outcomes and to investigate the interventions that effectively address the increased severity of PAD in dual-eligible patients.

Oncocytoma managed by active surveillance in a transplant allograft kidney: a case report.

BACKGROUND: The ethical implications of the utilization of kidneys with solid renal masses (SRMs) in transplantation are the subject of lively debate in the transplantation community and beyond. One of such implications is that as the life expectancy of renal transplant patients improve, the prevalence of SRMs in donors is likely to increase. We report a case of an oncocytoma in a renal allograft complicating a deceased-donor kidney transplant. CASE PRESENTATION: A 60-year-old woman received and underwent deceased-donor renal transplantation for end-stage renal disease after a waiting-list period of 11 years. Kidney Doppler ultrasound (DUS) of the deceased donor was negative for any nodular lesion. The finding of the DUS, done on postoperative day 1, to assess the patency of the graft, was suspicious for an acute arterial thrombosis but did not reveal any focal irregularities. An ensuing computed tomography (CT) scan did not show any arterial complications but serendipitously revealed a 2.4-cm lesion on the upper pole of the renal allograft, which was not detected during the back-table or ultrasonography monitoring. Histology of the biopsied lesion was consistent with oncocytoma. However, because the eosinophilic variant of chromophobe renal cell carcinoma may morphologically resemble renal oncocytoma, immunohistochemical staining was performed. The results were negative, ruling out chromophobe RCC. After discussing the therapeutic options and potential related outcomes with the patient, we found no reason for resection of the lesion or an allograft nephrectomy, given the low risk of malignant transformation in an oncocytoma. Active surveillance of the benign tumor was done with ultrasonography, every 2 months, for the first year and, then, with magnetic resonance imaging, every year. The patient received mycophenolate-mofetil, tacrolimus, and prednisone throughout the 5-year follow-up period, and the regimen for immunosuppression was not changed despite the presence of the renal mass. After 60 months, we report that none of the radiological findings have shown any morphological changes of the lesion, and the patient is well. CONCLUSION: To the best of our knowledge, we report the first case of an oncocytoma in a renal allograft complicating a deceased-donor kidney transplant, which was successfully managed by active surveillance.

Dual-eligible, dual-risk? A brief review on the impact of dual-eligible status on health disparities and peripheral artery disease.

Peripheral artery disease (PAD) has been associated with poorer outcomes based on particular social determinants of health, including insurance status. A unique population to study treatment outcomes related to PAD is those with dual-eligible status-those who qualify for both Medicare and Medicaid-comprising more than 12 million people. We performed a systematic review of the literature surrounding dual-eligible patients and impact on PAD, with final inclusion of six articles. Dual eligibility has been associated with higher rates of comorbidities; more severe symptoms at initial presentation for PAD; and poorer treatment outcomes, including mortality. Further studies are needed to specifically look at the association between PAD and dual-eligible status, but what is clear is that patients in this population would benefit from early identification to prevent disease progression and improve equity.

Laser speckle contrast imaging compared with indocyanine green in renal perfusion of a porcine model.

Background: When viewed under near-infrared light, indocyanine green (ICG) signal for kidney perfusion can be utilized in partial nephrectomy. Laser speckle contrast imaging (LSCI) uses coherent light to detect perfusion during real-time laparoscopic surgery. Materials and methods: Laser speckle contrast imaging or ActivSight, an imaging sensor adapter, was used during laparoscopy of an anesthetized porcine kidney model. ActivSight's "perfusion mode" and "quantification mode" displayed the blood flow as a heatmap and numerical signal intensity, respectively. Results: After the upper segmental renal artery was clamped, ICG was seen in the lower pole, and LSCI showed low unit (dark color) quantification and perfusion in the upper pole. Indocyanine green was retained in the lower pole after the upper segmental artery was unclamped, and LSCI perfusion was demonstrated in the entire kidney. Conclusions: Laser speckle contrast imaging is a dye-free, repeatable, real-time adjunct for renal parenchymal perfusion assessment applicable to minimally invasive renal surgery to complement the technology of ICG near-infrared fluorescence and advance digital surgery.

The Insurance Coverage Paradox - Characterizing Outcomes among Dual-Eligible Hemorrhagic Stroke Patients.

BACKGROUND: Socioeconomic factors, such as insurance status, have been shown to affect outcomes for patients following emergency injuries. Dual-eligible beneficiaries, receiving both Medicare and Medicaid, constitute an especially vulnerable population. There is limited data addressing whether dual-eligible beneficiaries with hemorrhagic stroke display unique characteristics and outcomes compared to patients with Medicare, Medicaid, or private insurance. STUDY DESIGN: We conducted a retrospective analysis of 10-years of National Inpatient Sample data. Using ICD-9-CM codes, we identified adult patients with known insurance status who were emergently hospitalized for intracranial hemorrhage; epidural, subdural, subarachnoid, and intracerebral hemorrhages were included. Patient characteristics including whether they underwent surgical intervention were collected. Multivariable logistic regression was used to adjust for confounders. Primary clinical outcomes of interest included mortality (in-hospital), complications (any), and favorable discharge (home/home with services). RESULTS: Among 410,621 patients, dual-eligible (6.8%) patients were on average older (mean age = 73yrs) compared to Medicaid (46yrs), private insurance (67yrs), or no-charge (47yrs) patients. Caucasian race was highest among Medicare patients (83%) while African-American race was highest among Medicaid (22%). Among all patients, 5.3% underwent operative intervention. Dual-eligibles had significantly higher odds of in-hospital mortality compared to no-charge (adjusted odds ratio (aOR) = 1.61, 95% CI = [1.04 - 2.49]), but no significant difference between Medicare and Medicaid although dual-eligibles. Dual-eligibles had significantly increased odds of complications compared to Medicaid (aOR = 1.23, 95% CI = [1.11 - 1.37]) and privately insured patients (aOR = 1.19, 95% CI = [1.11 - 1.28]), both p < 0.001, and lower odds of favorable discharge compared to all other groups, all p < 0.001. Dual-eligibles underwent a shorter length of stay, an 18% decrease, compared to Medicaid patients (ß-Coefficient = 0.82, 95% CI = [0.78 - 0.86], p < 0.001), and inflation adjusted admission costs that were 24% lower compared to Medicaid patients (ß-Coefficient = 0.76, 95% CI = [0.73 - 0.80], p < 0.001), amounting to a $3,684 decrease in cost. CONCLUSIONS: Dual-eligible beneficiaries experience unique health disparities from lower odds of favorable discharge to increased odds of complications and in-hospital mortality compared to other insured and uninsured groups. Adverse outcomes among dual-eligible beneficiaries highlight the need to uncover and address unknown sources of disparities to improve emergency treatment of hemorrhagic stroke in this population.

Providing Appropriate Pancreatic Cancer Care for People Experiencing Homelessness: A Surgical Perspective.

People experiencing homelessness are particularly vulnerable when diagnosed with pancreatic cancer. Patients with lower socioeconomic status have worse outcomes from pancreatic cancer as the result of disparities in access to treatment and barriers to navigation of the health care system. Patients with lower socioeconomic status, or who are vulnerably housed, are less likely to receive surgical treatment even when it is recommended by National Comprehensive Cancer Network guidelines. This disparity in access to surgical care explains much of the gap in pancreatic cancer outcomes. There are many factors that contribute to this disparity in surgical management of pancreatic cancer in people experiencing homelessness. These include a lack of reliable transportation, feeling unwelcome in the medical setting, a lack of primary care and health insurance, and implicit biases of health care providers, including racial bias. Solutions that focus on rectifying these problems include utilizing patient navigators, addressing implicit biases of all health care providers and staff, creating an environment that caters to the needs of patients experiencing homelessness, and improving their access to insurance and regional support networks. Implementing these potential solutions all the way from the individual provider to national safety nets could improve outcomes for patients with pancreatic cancer who are experiencing homelessness.

Multicenter Randomized Controlled Trial to Assess an e-Learning on Acute Burns Management.

An e-learning tutorial was developed to cover the basic aspects of acute burns management. The aim of this study was to provide objective educational evidence supporting the role of an e-learning on acute burns management ("basic burns management" or BBM) when compared with traditional teaching methods when introduced in different medical school settings around the world. A multicenter randomized controlled trial was conducted at the University of Pittsburgh, Peninsula College of Medicine and Dentistry, and St. George's University of London comparing the learning experience of medical students with the BBM e-learning tool (intervention) and a traditional lecture (control). A group of medical students was randomly allocated to the e-learning or lecture arms. Both groups were subjected to a 10-question pre-intervention and post-intervention tests assessing burns knowledge, and were asked to fill out a satisfaction survey. A total of 79 medical students of varying years of study participated. As a whole, students demonstrated a significant gain in knowledge after intervention (overall = 47.6%, P < .001), regardless of medical school year of study or interest in surgery. Participants undertaking the BBM e-learning had a greater exam score improvement and satisfaction compared with the traditional lecture even though they were not statistically significant. BBM e-learning is a free tool that provides comparable acute burn care learning opportunity and satisfaction outcomes to a traditional lecture, allowing convenient and standardized incorporation of burns teaching within an educational setting, regardless of geographical location, level of experience, or interest in surgery.

Seroconversion rates among health care workers exposed to hepatitis C virus-contaminated body fluids: The University of Pittsburgh 13-year experience.

BACKGROUND: Hepatitis C virus (HCV) transmission to health care personnel (HCP) after exposure to a HCV-positive source has been reported to occur at an average rate of 1.8% (range, 0%-10%). We aimed to determine the seroconversion rate after exposure to HCV-contaminated body fluid in a major U.S. academic medical center. METHODS: A longitudinal analysis of a prospectively maintained database of reported occupational injuries occurring between 2002 and 2015 at the University of Pittsburgh Medical Center was performed. Data collected include type of injury and fluid, injured body part, contamination of sharps, resident physicians' involvement, and patients' hepatitis B virus (HBV), HCV, and HIV status. RESULTS: A total of 1,361 cases were included in the study. Most exposures were caused by percutaneous injuries (65.0%), followed by mucocutaneous injuries (33.7%). Most (63.3%) were injuries to the hand, followed by the face and neck (27.6%). Blood exposure accounted for 72.7%, and blood-containing saliva accounted for 3.4%. A total of 6.9% and 3.7% of source patients were coinfected with HIV and HBV, respectively. The HCV seroconversion rate was 0.1% (n = 2) because of blood exposure secondary to percutaneous injuries. CONCLUSIONS: This study provides the largest and most recent cohort from a major U.S. academic medical center. The seroconversion rates among HCP exposed to HCV-contaminated body fluids was found to be lower than most of the data found in the literature.

Seroconversion rate among health care workers exposed to HIV-contaminated body fluids: The University of Pittsburgh 13-year experience.

BACKGROUND: The studies enumerating the risk of HIV transmission to health care workers (HCWs) as 0.3% after percutaneous exposure to HIV-positive blood, and 0.09% after a mucous membrane exposure, are weakened by dated literature. Our study aims to demonstrate the seroconversion rate after exposure to HIV-contaminated body fluids in a major academic center in the United States. METHODS: A prospectively maintained database of reported occupational injuries occurring between 2002 and 2015 at an academic medical center was analyzed. Data collected included the type of injury, injured body part, type of fluid, contamination of sharps, involvement of resident physicians, use of postexposure prophylaxis, and patients' HIV, hepatitis B virus, and hepatitis C virus status. RESULTS: A total of 266 cases were included in the study. Most exposures were caused by percutaneous injuries (52.6%), followed by 43.2% mucocutaneous injuries. Of the injuries, 52.6% were to the hand and 33.5% to the face and neck. Blood exposure accounted for 64.3% of all cases. Of the patients, 21.1% received postexposure prophylaxis. None of the HCWs exposed to HIV-contaminated body fluids seroconverted (seroconversion rate, 0%). CONCLUSIONS: HIV does not seem to be as easily transmitted by needlestick, laceration, or splash injuries as previously surmised. Further large-scale and multicenter studies are needed for a more accurate estimation of the risk of transmission of HIV in U.S. health care workers.