Factors Associated With Pelvic Infection After Pre-Peritoneal Pelvic Packing for Hemodynamically Unstable Pelvic Fractures.

Background: Preperitoneal pelvic packing (PPP) and external fixation has led to improved mortality after devastating pelvic trauma. However, there is limited literature on infection after this intervention. We aim to study the risk factors associated with pelvic infection after PPP. Patients and Methods: A retrospective review of patients who underwent PPP at a single level 1 trauma center was performed. Results: Over the 18-year study period, 222 patients were identified. Twenty-three percent of patients had an open fracture. Pelvic angiography was performed in 24% of patients with 16% requiring angioembolization (AE). The average time to packing removal was two (one to two days) days, although 10% of patients had their pelvis re-packed. Overall infection rate was 14% (n = 31); if pelvic re-packing was performed, the infection rate increased to 45%. Twenty-two of the patients with an infection required additional procedures for their infection, and ultimately hardware removal occurred in eight patients. On univariable analysis, patients with pelvic infections had more open fractures (55% vs. 17%; p < 0.01), underwent AE more frequently (29% vs. 14%; p = 0.04), were more likely to undergo repacking (32% vs. 6%; p < 0.01), and had packing in place for longer (2 [1,2] vs. 2 [2,3]; p = 0.01). On logistic multivariable regression analysis, open fracture (odds ratio [OR], 5.8; 95% confidence interval [CI], 2.4-14.1) and pelvic re-packing (OR, 4.7; 95% CI, 1.2-18.5) were independent risk factors for pelvic infection. Conclusions: Pelvic infection after PPP is a serious complication independently associated with open fracture and re-packing of the pelvis. Re-intervention was required in most patients with infection.

Management of Pelvic Trauma.

Pelvic fractures are common after blunt trauma with patients' presentation ranging from stable with insignificant fractures to life-threatening exsanguination from unstable fractures. Often, hemorrhagic shock from a pelvic fracture may go unrecognized and high clinical suspicion for a pelvic source lies with the clinician. A multidisciplinary coordinated effort is required for management of these complex patients. In the exsanguinating patient, hemorrhage control remains the top priority and may be achieved with external stabilization, resuscitative endovascular balloon occlusion of the aorta, preperitoneal pelvic packing, angiographic intervention, or a combination of therapies. These modalities have been shown to reduce mortality in this challenging population.

How Informed Is Your Informed Consent: Evaluating Differences Between Resident and Attending Obtained Consents for Cholecystectomy.

OBJECTIVE: There is considerable variability in surgeons' approach to write and obtain informed consent for surgery, particularly among resident trainees. We analyzed differences in procedures and complications described in documented surgical consents for cholecystectomy between residents and attendings. We hypothesized that attending consents would describe more comprehensive procedures and complications than those done by residents. DESIGN: This is a retrospective analysis of 334 patients who underwent cholecystectomy. Charts were queried for demographics, surgical approach, whether the consent was completed electronically, and which provider completed the consent. Specifically, consents were evaluated for inclusion of possible conversion to open procedure, intraoperative cholangiogram, bile duct injury, injury to nearby structures, reoperation, bile leak, as well as if the consent matched the actual procedure performed. SETTING: This study was conducted at an accredited general surgery training program at an academic tertiary care center in the Midwest. PARTICIPANTS: This was a review of 334 patients who underwent cholecystectomy over a 1 year period. RESULTS: Of all documented consents analyzed, 153 (47%) specifically included possible intraoperative cholangiogram, 156 (47%) included bile duct injury, 76 (23%) included injury to nearby structures, 22 (7%) included reoperation, and 62 (19%) included bile leak. In comparing residents and attendings, residents were more likely to consent for bile duct injury (p = 0.002), possible intraoperative cholangiogram (p = 0.0007), injury to nearby structures (p < 0.0001), reoperation (p < 0.0001), and bile leak (p < 0.0001). CONCLUSIONS: Significant variation exists between documentation between resident and attending cholecystectomy consents, with residents including more complications than attendings on their consent forms. These data suggest that experience alone does not predict content of written consents, particularly for common ambulatory procedures. Education regarding the purpose of informed consent and what should be included in one may lead to a reduction in variability between providers.

Antimicrobial Coating Prevents Ventilator-Associated Pneumonia in a 72 hour Large Animal Model.

BACKGROUND: The primary goal of this study was to demonstrate that endotracheal tubes coated with antimicrobial lipids plus mucolytic or antimicrobial lipids with antibiotics plus mucolytic would significantly reduce pneumonia in the lungs of pigs after 72 hours of continuous mechanical ventilation compared to uncoated controls. MATERIALS AND METHODS: Eighteen female pigs were mechanically ventilated for up to 72 hours through uncoated endotracheal tubes, endotracheal tubes coated with the antimicrobial lipid, octadecylamine, and the mucolytic, N-acetylcysteine, or tubes coated with octadecylamine, N-acetylcysteine, doxycycline, and levofloxacin (6 pigs per group). No exogenous bacteria were inoculated into the pigs, pneumonia resulted from the pigs' endogenous oral flora. Vital signs were recorded every 15 minutes and arterial blood gas measurements were obtained for the duration of the experiment. Pigs were sacrificed either after completion of 72 hours of mechanical ventilation or just prior to hypoxic arrest. Lungs, trachea, and endotracheal tubes were harvested for analysis to include bacterial counts of lung, trachea, and endotracheal tubes, lung wet and dry weights, and lung tissue for histology. RESULTS: Pigs ventilated with coated endotracheal tubes were less hypoxic, had less bacterial colonization of the lungs, and survived significantly longer than pigs ventilated with uncoated tubes. Octadecylamine-N-acetylcysteine-doxycycline-levofloxacin coated endotracheal tubes had less bacterial colonization than uncoated or octadecylamine-N-acetylcysteine coated tubes. CONCLUSION: Endotracheal tubes coated with antimicrobial lipids plus mucolytic and antimicrobial lipids with antibiotics plus mucolytic reduced bacterial colonization of pig lungs after prolonged mechanical ventilation and may be an effective strategy to reduce ventilator-associated pneumonia.

Attending and Resident Surgeon Perspectives and Prescribing Practices of Pain Medication During the Opioid Epidemic.

OBJECTIVE: Over 67,000 individuals died in the United States due to drug overdose in 2018; the majority of these deaths were secondary to opioid ingestion. Our aim was to determine surgeon perceptions on opioid abuse, the adequacy of perioperative and graduate medical education, and the role surgeons may play. We also aimed to investigate any differences in attending and resident surgeon attitudes. DESIGN: Anonymous online survey assessing surgeons' opioid counseling practices, prescribing patterns, and perceptions on opioid abuse, adequacy of education about opioid abuse, and the role physicians play. SETTING: Two Accreditation Council for Graduate Medical Education accredited general surgery programs at a university-based tertiary hospital and a community hospital in the Midwest. PARTICIPANTS: Attending and resident physicians within the Departments of Surgery participated anonymously. RESULTS: Attending surgeons were more likely than residents to discuss posoperative opioids with patients (62% vs. 33%; p < 0.05), discuss the potential of opioid abuse (31% vs. 6%; p < 0.05), and check state-specific prescription monitoring programs (15% vs. 0%; p < 0.05). Surgeons and trainees feel that surgeons have contributed to the opioid epidemic (76% attending vs. 88% resident). Overall, attending and resident surgeons disagree that there is adequate formal education (66% vs. 66%) but adequate informal education (48% vs. 61%) on opioid prescribing. However, when attending physicians were broken down into those who have practiced ≤5 years vs. those with >5 years experience, those with ≤5 years experience were more confident in recognizing opioid abuse (61% vs. 34%) and fewer young faculty disagreed that there is adequate formalized education on opioid prescribing (45% vs. 84%). CONCLUSION AND RELEVANCE: Patient education should be improved upon in the preoperative setting and should be treated as an important component of preoperative discussions. Formalized opioid education should also be undertaken in graduate surgical education to help guide appropriate opioid use by resident and attending physicians.

Doxycycline-Coated Silicone Breast Implants Reduce Acute Surgical-Site Infection and Inflammation.

BACKGROUND: Surgical-site infection after implant-based breast reconstruction remains a leading cause of morbidity. Doxycycline is an antibiotic used to treat soft-tissue infections. The authors hypothesize that doxycycline-coated breast implants will significantly reduce biofilm formation, surgical-site infection, and inflammation after bacterial infection. METHODS: Pieces of silicone breast implants were coated in doxycycline. In vitro studies to characterize the coating include Fourier transmission infrared spectroscopy, elution data, and toxicity assays (n = 4). To evaluate antimicrobial properties, coated implants were studied after methicillin-resistant Staphylococcus aureus (MRSA) and Pseudomonas aeruginosa inoculation in vitro and in a mouse model at 3 and 7 days (n = 8). Studies included bacterial quantification, cytokine profiles, and histology. RESULTS: Coated silicone breast implants demonstrated a color change, increased mass, and Fourier transmission infrared spectroscopy consistent with a doxycycline coating. Coated implants were nontoxic to fibroblasts and inhibited biofilm formation and bacterial adherence after MRSA and P. aeruginosa incubation in vitro, and measurable doxycycline concentrations at 24 hours were seen. In a mouse model, a significant reduction of MRSA and P. aeruginosa bacterial colonization after 3 and 7 days in the doxycycline-coated implant mice was demonstrated when compared to the control mice, control mice treated with intraperitoneal doxycycline, and control mice treated with a gentamicin/cefazolin/bacitracin wash. Decreased inflammatory cytokines and inflammatory cell infiltration were demonstrated in the doxycycline-coated mice. CONCLUSIONS: A method to coat silicone implants with doxycycline was developed. The authors' doxycycline-coated silicone implants significantly reduced biofilm formation, surgical-site infections, and inflammation. Further studies are needed to evaluate the long-term implications.

Swipe right for surgical residency: Exploring the unconscious bias in resident selection.

BACKGROUND: Applicants provide a photo with their application through the Electronic Residency Application Service, which may introduce appearance-based bias. We evaluated whether an unconscious appearance bias exists in surgical resident selection. METHODS: After the match, applicant data from the 2018 to 2019 and 2019 to 2020 application cycles were examined. Reviewers were not provided the applicant photo or self-identified race during the second cycle. Photos provided by candidates were then rated by 4 surgical subspecialty residents who had no prior exposure to applications or interview status. Photos were rated on perceived fitness level, visual appearance, and photo professionalism. An overall photo score was then calculated. RESULTS: In the study, 422 applications were reviewed and 164 received interview invitations during the 2018 to 2019 cycle. Alpha Omega Alpha membership (odds ratio, 2.31; 95% confidence interval, 1.18-4.51), overall photo score (odds ratio, 2.29, 95% confidence interval, 1.43-3.66), research (odds ratio, 5.61, 95% confidence interval, 2.84-11.20), age (odds ratio, 0.86, 95% confidence interval, 0.76-0.99), and step 2 (odds ratio, 1.06, 95% confidence interval, 1.03-1.09) were predictors for receiving an interview. For the 2019 to 2020 cycle, 398 applications were reviewed, and 75 applicants received an invitation. Step 2 (odds ratio, 1.07, 95% confidence interval, 1.02-1.12), research (odds ratio, 2.78, 95% confidence interval, 1.40-5.55), age (odds ratio, 0.82, 95% confidence interval, 0.71-0.95), and overall photo score (odds ratio, 2.27; 95% confidence interval, 1.14-4.52) remained predictors despite reviewers being blinded to the photo during this cycle. CONCLUSION: Although objective metrics remain critical in determining interview invitations, overall perceived applicant appearance may influence the selection process. Although visual appearance was associated with receiving an interview, the Electronic Residency Application Service photo does not ultimately affect selection. This may suggest that appearance may influence other objective and subjective aspects of the application.

Does chest wall Organ Injury Scale (OIS) or Abbreviated Injury Scale (AIS) predict outcomes? An analysis of 16,000 consecutive rib fractures.

BACKGROUND: Chest wall injuries have serious clinical consequences. It is presumed a higher severity of injury correlates with worse outcomes. The 2 most common chest wall injury severity scores, the Organ Injury Scale and the Abbreviated Injury Scale, are based on expert opinion with unknown clinical endpoints. Our aim was to determine if either the Organ Injury Scale or the Abbreviated Injury Scale are associated with clinical outcomes. METHODS: A single institution, 4-year retrospective study of all patients with rib or sternal fractures was conducted. All patients were assessed for both Organ Injury Scale and Abbreviated Injury Scale scores. Outcomes assessed included mortality, complications, tracheostomy, and readmissions. Receiver operating characteristic areas under the curve were calculated to measure discriminatory accuracy of scoring systems for outcomes in chest wall injury. RESULTS: Overall, 3,033 patients presented with a total of 16,055 rib fractures. The median chest wall scores were 2 for Organ Injury Scale and 3 for Abbreviated Injury Scale. Abbreviated Injury Scale scores for the same patients were greater than the Organ Injury Scale in 48.7%, equivalent in 46.7%, and lower in 4.6%. The receiver operating characteristic areas under the curve for in-hospital outcomes were weakly predictive for the Organ Injury Scale over the Abbreviated Injury Scale. The receiver operating characteristic areas under the curve for readmissions were very weakly predictive for the Abbreviated Injury Scale over the Organ Injury Scale. CONCLUSION: There is a very weak association between chest wall Organ Injury Scale score and in-hospital outcomes. The Abbreviated Injury Scale score outperformed the Organ Injury Scale, only being weakly predictive of readmission. Chest wall injury scoring systems may need revision for future outcomes-based research and practice improvements.

Death by Decade: Establishing a Transfusion Ceiling for Futility in Massive Transfusion.

BACKGROUND: Age and massive transfusion are predictors of mortality after trauma. We hypothesized that increasing age and high-volume transfusion would result in progressively elevated mortality rates and that a transfusion "ceiling" would define futility. METHODS: The Trauma Quality Improvement Program (TQIP) database was queried for 2013-2016 records and our level I trauma registry was reviewed from 2013 to 2018. Demographic, mortality, and blood transfusion data were collected. Patients were grouped by decade of life and by packed red blood cell (pRBC) transfusion requirement (zero units, 1-3 units, or ≥4 units) within 4 h of admission. RESULTS: TQIP analysis demonstrated an in-hospital mortality risk that increased linearly with age, to an odds ratio of 10.1 in ≥80 y old (P < 0.01). Mortality rates were significantly higher in older adults (P < 0.01) and those with more pRBCs transfused. In massively transfused patients, the transfusion "ceiling" was dependent on age. Owing to the lack granularity in the TQIP database, 230 patients from our institution who received ≥4 units of pRBCs within 4 h of admission were reviewed. On arrival, younger patients had significantly higher heart rates and more severe derangements in lactate levels, base deficits, and pH compared with older patients. There were no differences among age groups in injury severity score, systolic blood pressure, or mortality. CONCLUSIONS: In massively transfused patients, mortality increased with age. However, a significant proportion of older adults were successfully resuscitated. Therefore, age alone should not be considered a contraindication to high-volume transfusion. Traditional physiologic and laboratory criteria indicative of hemorrhagic shock may have reduced reliability with increasing age, and thus providers must have a heightened suspicion for hemorrhage in the elderly. Early transfusion requirements can be combined with age to establish prognosis to define futility to help counsel families regarding mortality after traumatic injury.

Readmission rates and associated factors following rib cage injury.

BACKGROUND: There remains a lack of knowledge about readmission characteristics after sustaining rib fractures. We aimed to determine rates, characteristics, and predictive/protective factors associated with unexpected reevaluation and readmission after rib cage injury. METHODS: A retrospective review was performed based on trauma patients evaluated at an urban Level I trauma center from January 2014 to December 2016. Adult patients sustaining blunt trauma with more than one rib fracture or a sternomanubrial fracture were defined as having moderate to severe rib cage injury. Exclusion criteria included penetrating injury, death during initial hospitalization, and only one rib fracture. Reevaluation was defined as presenting at a hospital within 90 days of discharge urgently or emergently. Demographics, injury characteristics, comorbidities, complications, imaging, and readmission data were collected. Univariate and multivariate analysis was performed with a significance of p less than 0.05. RESULTS: During the study period, 11,667 patients underwent trauma evaluation, of which 1,717 patients were found to have a moderate to severe rib cage injury. Within 90 days, 397 (23.1%) of patients underwent reevaluation, while 177 (10.3%) required readmission. One hundred forty-two (8.3%) patients were reevaluated specifically for chest-related complaints, and 55 (3.2%) required readmission. On univariate analysis, Injury Severity Score greater than 15, hospital length of stay longer than 7 days, intensive care unit (ICU) length of stay longer than 3 days, a worsened chest x-ray at discharge, a psychiatric comorbidity, a smoking comorbidity, deep vein thrombosis, unplanned readmission to the ICU, and unplanned intubation were higher in the overall reevaluation cohort. On multivariate analysis, age of 15 years to 35 years, Risk Assessment Profile score greater than 8, hypertension, psychiatric comorbidity, current smoker, and unplanned return to the ICU on index admission were predictive of reevaluation of overall reevaluation. CONCLUSION: Moderate to severe rib cage injury is associated with high rates of reevaluation and readmission. Younger patients who smoke and required a return to the ICU are at greater risk for readmission. LEVEL OF EVIDENCE: Level IV, Prognostic and Epidemiologic.

Optimizing Lower Extremity Duplex Ultrasound Screening After Traumatic Injury.

BACKGROUND: The risk assessment profile (RAP) score has been used to determine patients who would most benefit from lower extremity duplex ultrasound screening (LEDUS). We hypothesized that revising our LEDUS protocol to perform screening ultrasound examinations in patients with an RAP ≥8 within 48 h of admission would reduce the number of LEDUS performed without changing outcomes. METHODS: A retrospective review was conducted on trauma patients admitted from July 1, 2014, to June 30, 2015, and July 1, 2016, to June 30, 2017. In 2014-2015, patients with an RAP score ≥5 underwent weekly LEDUS examinations starting on hospital day 4. In 2016-2017, the protocol was changed to start screening patients with an RAP score ≥8 by hospital day 2. Both protocols screened with weekly ultrasounds after the first examination. Demographic data, injury characteristics, LEDUS examination findings, chemoprophylaxis type, and venous thromboembolism incidence were collected. RESULTS: A total of 602 patients underwent LEDUS examination in 2014-2015, whereas only 412 underwent LEDUS in 2016-2017. No significant difference was seen in the number of patients diagnosed with deep vein thrombosis (DVT) or pulmonary embolism. DVTs were most often identified on the first LEDUS examination in both cohorts. Of patients diagnosed with a DVT on an LEDUS examination, a significantly higher RAP score (12 versus 10), and a shorter time to first duplex (1 versus 3 d), and DVT diagnosis (2 versus 4 d) were observed in the 2016-2017 cohort. In patients diagnosed with a pulmonary embolism, no significant differences were demonstrated between cohorts. CONCLUSIONS: Refinement of LEDUS protocols can decrease overutilization of hospital resources without compromising trauma patient outcomes.

Surgeons, Scholars, and Leaders Symposium: A 5-Year Experience.

The Department of Surgery at the University of Cincinnati developed the Surgeons, Scholars, and Leaders Symposium to address the underappreciated aspects of surgical education that are critical in the development of the academic surgeon. Surgical education has undergone many gaps since the beginning of a traditional surgical residency, first pioneered by Dr. Halsted in 1904; still, many gaps in surgical education remain. Topics such as research, financial planning, leadership, career development, and many others are not adequately addressed in formalized training. The Surgeons, Scholars, and Leaders Symposium was first held in January 2015 in Jackson Hole, WY, and has subsequently become an annual event. Recurrent themes addressed at the Symposium include global health, resident autonomy, research program development, leadership, mentorship, career development, and managing transitions. The annual Surgeons, Scholars, and Leaders Symposium has been instrumental in addressing these underappreciated aspects of surgeon development and will continue to be an important venue for the next generation of surgical leaders.

Evaluation of a Novel Fibrin Sealant Patch in Hemorrhage Control After Vascular or Hepatic Injury.

INTRODUCTION: Acute hemorrhage remains the leading cause of death in potentially survivable injuries. The use of topical hemostatic agents has increased over the last two decades with the evolution of damage control surgery. By 2008, the military widely adopted Combat Gauze as the hemostatic dressing of choice for compressible hemorrhage. The goal of this study was to compare the performance of a novel fibrin sealant patch to Combat Gauze in two clinically relevant models of hemorrhage. MATERIALS AND METHODS: Yorkshire swine underwent unilateral femoral artery puncture or a grade V liver laceration with timed free bleeding then received either the fibrin patch or Combat Gauze packing with 3 minutes of standardized pressure. Animals were then resuscitated to maintain a mean arterial pressure of 60 mmHg for 4 hours. Hemostasis, blood loss, resuscitation volume, survival, vessel patency, and hematologic parameters were evaluated. RESULTS: Hemostasis was equivalent in both groups after hepatic and vascular injury. Survival was 80% in the fibrin patch vascular injury group and 100% in all other groups. Hematologic parameters were not significantly different between treatment groups. Femoral artery patency was 80% in both groups after vascular injury. With simulated ambulation after vessel injury, 60% of the Combat Gauze group and 80% of the fibrin patch group remained hemostatic (p > 0.05). In simulated re-exploration with packing removal, all animals rebled after hemostatic product removal. CONCLUSION: There was no significant difference in hemostasis between a novel fibrin patch and Combat Gauze after extremity arterial or hepatic injury. This novel fibrin patch may have a clinical advantage over the Combat Gauze, as it can be left in the body, thereby limiting the potential need for reoperation.

Sphingolipids and Innate Immunity: A New Approach to Infection in the Post-Antibiotic Era?

Antibiotic resistance has been demonstrated during the entire duration of antibiotic use even before medical utilization. Increasing resistance within virtually all microbes continues to be a problem. Infection with antibiotic resistant microbes has demonstrated significantly increased morbidity, death, and health-care-associated costs. Given increasing antibiotic resistance, multiple novel agents and approaches are being investigated, including antimicrobial lipids. Sphingosine and ceramide have been demonstrated to play a pivotal role in the innate immunity of the epidermis, oral mucosa, and respiratory epithelium; their role is being investigated currently in uroepithelium. Ceramide has been shown to be pivotal in the regulation of mammalian defense against Pseudomonas aeruginosa and Staphylococcus aureus pathogens commonly encountered in pneumonia. On the other hand, sphingosine appears to be equally pivotal and directly involved in pathogenic defense and has been demonstrated to "rescue" mammals from P. aeruginosa infections. Within this review, we will discuss the role of sphingolipids within innate immunity, pathogen invasion, and bacterial infection. We will discuss the antimicrobial activity of sphingosine and possibility for commercial use as an antimicrobial in the post-antibiotic era.

Variability of fluid administration during exploratory laparotomy for abdominal trauma.

BACKGROUND: Approximately 8% of traumatically injured patients require transfusion with packed red blood cells (pRBC) and only 1% to 2% require massive transfusion. Intraoperative massive transfusion was defined as requiring greater than 5 units (u) of pRBC in 4 hours. Despite the majority of patients not requiring transfusion, the appropriate amount and type of crystalloid administered during the era of damage control resuscitation have not been analyzed. We sought to determine the types of crystalloid used during trauma laparotomies and the potential effects on resuscitation. METHODS: Patients who underwent laparotomy after abdominal trauma from January 2014 to December 2016 at the University of Cincinnati Medical Center were identified. Patients were grouped based on requiring 0u, 1u to 4u, and ≥5u pRBC during intraoperative resuscitation. Demographic, physiologic, pharmacologic, operative, and postoperative data were collected. Statistical analysis was performed with Kruskal-Wallis test and Pearson's correlation coefficient. RESULTS: Lactated Ringer's (LR) solution was the most used crystalloid type received in the 0u and 1u to 4u pRBC cohorts, whereas normal saline (NS) was the most common in the ≥5u pRBC cohort. Most patients received two types of crystalloid intraoperatively. NS and LR were most frequently the first crystalloids administered, with Normosol infusion occurring later. The amount of crystalloid received correlated with operative length, but did not correlate with the estimated blood loss. Neither the type of crystalloid administered nor the anesthesia provider type was associated with changes in postoperative resuscitation parameters or electrolyte concentrations. DISCUSSION: There is a wide variation in the amount and types of crystalloids administered during exploratory laparotomy for trauma. Interestingly, the amount or type of crystalloid given did not affect resuscitation parameters regardless of blood product requirement. LEVEL OF EVIDENCE: Level IV.

Kinetics of engraftment in patients with hematologic malignancies given allogeneic hematopoietic cell transplantation after nonmyeloablative conditioning.

We analyzed the kinetics of donor engraftment among various peripheral blood cell subpopulations and their relationship to outcomes among 120 patients with hematologic malignancies given hematopoietic cell transplantation (HCT) after nonmyeloablative conditioning consisting of 2 Gy total body irradiation (TBI) with or without added fludarabine. While patients rapidly developed high degrees of donor engraftment, most remained mixed donor/host chimeras for up to 180 days after HCT. Patients given preceding chemotherapies and those given granulocyte colony-stimulating factor-mobilized peripheral blood mononuclear cell (G-PBMC) grafts had the highest degrees of donor chimerism. Low donor T-cell (P = .003) and natural killer (NK) cell (P = .004) chimerism levels on day 14 were associated with increased probabilities of graft rejection. High T-cell chimerism on day 28 was associated with an increased probability of acute graft-versus-host disease (GVHD) (P = .02). Of 93 patients with measurable malignant disease at transplantation, 41 achieved complete remissions a median of 199 days after HCT; 19 of the 41 were mixed T-cell chimeras when complete remissions were achieved. Earlier establishment of donor NK-cell chimerism was associated with improved progression-free survival (P = .02). Measuring the levels of peripheral blood cell subset donor chimerisms provided useful information on HCT outcomes and might allow early therapeutic interventions to prevent graft rejection or disease progression.