Faculty Bibliography

Induction therapy with rabbit anti-thymocyte globulin (rATG) in low-risk kidney transplant recipients (KTR) remains controversial, given the associated increased risk of cytomegalovirus (CMV) infection. This natural experiment compared 12-month clinical outcomes in low-risk KTR without CMV prophylaxis (January/3/13-September/16/15) receiving no induction or a single 3 mg/kg dose of rATG. We used logistic regression to characterize delayed graft function (DGF), negative binomial to characterize length of hospital stay (LOS), and Cox regression to characterize acute rejection (AR), CMV infection, graft loss, death, and hospital readmissions. Recipients receiving 3 mg/kg rATG had an 81% lower risk of AR (aHR _0.14 0.19_0.25 , P < 0.001) but no increased rate of hospital readmissions because of infections (_0.68 0.91_1.21 , P = 0.5). There was no association between 3 mg/kg rATG and CMV infection/disease (aHR _0.86 1.10_1.40 , P = 0.5), even when the analysis was stratified according to recipient CMV serostatus positive (aHR _0.94 1.25_1.65 , P = 0.1) and negative (aHR _0.28 0.57_1.16 , P = 0.1). There was no association between 3 mg/kg rATG and mortality (aHR _0.51 1.25_3.08 , P = 0.6), and graft loss (aHR _0.34 0.73_1.55 , P = 0.4). Among low-risk KTR receiving no CMV pharmacological prophylaxis, 3 mg/kg rATG induction was associated with a significant reduction in the incidence of AR without an increased risk of CMV infection, regardless of recipient pretransplant CMV serostatus.

BACKGROUND:The aim of this study is to describe the economic trends in adults who underwent elective thyroidectomy. METHODS:We performed a population-based study utilizing the Premier Healthcare Database to examine adult patients who underwent elective thyroidectomy between January 2006 and December 2014. Time was divided into three equal time periods (2006-2008, 2009-2011, and 2012-2014). To examine trend in patient charges, we modeled patient charges using generalized linear regressions adjusting for key covariates with standard errors clustered at the hospital level. RESULTS:Our study cohort consisted of 52,012 adult patients who underwent a thyroid operation. During the study period, the most common procedure changed from a thyroid lobectomy to bilateral thyroidectomy. Over the study period, there was an increase in the proportion of completion thyroidectomies from 1.1% to 1.6% (P < 0.001), malignant diagnoses from 21.7% to 26.8% (P < 0.001), procedures performed at teaching hospitals from 27.7% to 32.9% (P < 0.001), and procedures performed on an outpatient basis from 93.85% to 97.55% (P < 0.001). The annual increase in median patient charge adjusted for inflation was $895 or 4.3% resulting in an increase of 38.8% over 9 y. Higher thyroidectomy charges were associated with male patients, malignant surgical pathology, patients undergoing limited or radical neck dissection, experiencing complications, those with managed health care insurance, and a prolonged length of stay. CONCLUSIONS:Despite recent changes in thyroid surgery practices to decrease the economic burden of hospitals, costs continue to rise 4.3% annually. Additional prospective studies are needed to identify factors associated with this increasing cost.

BACKGROUND AND AIMS:Scores from the Model for End-Stage Liver Disease (MELD), which are used to prioritize candidates for deceased donor livers, are widely acknowledged to be negatively correlated with the 90-day survival rate without a liver transplant. However, inconsistent and outdated estimates of survival probabilities by MELD preclude useful applications of the MELD score. APPROACH AND RESULTS:Using data from all prevalent liver waitlist candidates from 2016 to 2019, we estimated 3-day, 7-day, 14-day, 30-day, and 90-day without-transplant survival probabilities (with confidence intervals) for each MELD score and status 1A. We used an adjusted Kaplan-Meier model to avoid unrealistic assumptions and multiple observations per person instead of just the observation at listing. We found that 90-day without-transplant survival has improved over the last decade, with survival rates increasing >10% (in absolute terms) for some MELD scores. We demonstrated that MELD correctly prioritizes candidates in terms of without-transplant survival probability but that status 1A candidates' short-term without-transplant survival is higher than that of MELD 40 candidates and lower than that of MELD 39 candidates. Our primary result is the updated survival functions themselves. CONCLUSIONS:We calculated without-transplant survival probabilities for each MELD score (and status 1A). The survival function is an invaluable tool for many applications in liver transplantation: awarding of exception points, calculating the relative demand for deceased donor livers in different geographic areas, calibrating the pediatric end-stage liver disease score, and deciding whether to accept an offered liver.

BACKGROUND:clinical and imaging investigations allow a detailed assessment of an organ donor, but a quota of cancer still elude detection. Complete autopsy of donors is even less frequently performed, due to economic issues and increasing availability of high-quality imaging. The aim of this study is to gather evidence from the literature on donor malignancy discovered at autopsy following organ donation and to discuss the utility and limitations of autopsy practice in the field of transplantation. METHODS:A systematic search according to PRISMA guidelines was carried out in Pubmed and Embase databases until September 2020 to select articles with reporting of cancer discovered in a donor at postmortem examination. Cancer discover in not-transplant setting were excluded. A descriptive synthesis was provided. RESULTS:Of 7388 articles after duplicates removal, 56 were included. Fifty-one studies reported on complete autopsy, while 5 dealt only with limited autopsy (prostate and central nervous system). The number of autopsies ranged between 1 and 246 with a total of 823 autopsies performed. The most frequent cancer discovered at autopsy was lymphoma (n = 13, 15%), followed by renal cell carcinoma (RCC) (n = 11, 13%), non-small cell lung cancer (NSCLC) (n = 10, 11%), melanoma (n = 10, 11%), choriocarcinoma (n = 6, 7%) and glioblastoma (GBM) (n = 6, 7%). CONCLUSIONS:Lymphoma and melanoma are still difficult-to-detect cancers both during donor investigation and at procurement, whilst prostate cancer and choriocarcinoma are almost always easily detected nowadays thank to blood markers and clinical examination. There have been improvements with time in pre-donation detection procedures which are now working well, particularly when complete imaging investigations are performed, given that detection rate of CT/MRI is high and accurate. Autopsy can play a role to help to establish the correct donor management pathways in case of cancer discover. Furthermore, it helps to better understand which cancers are still eluding detection and consequently to refine guidelines' assessment procedures.

BACKGROUND:Individuals undergoing hemodialysis in the United States frequently report pain and receive three-fold more opioid prescriptions than the general population. While opioid use is appropriate for select patients, high-dose utilization may contribute to an increased risk of death due to possible accumulation of opioid metabolites. METHODS:We studied high-dose opioid utilization (≥120 morphine milligram equivalents [MME] per day) among adults initiating hemodialysis in the United States between 2007 and 2014 using national registry data. We calculated the cumulative incidence (%) of high-dose utilization and depicted trends in the average percentage of days individuals were exposed to opioids. We used adjusted Cox proportional hazards models to identify which opioid doses were associated with mortality. RESULTS:Among 327,344 adults undergoing hemodialysis, the cumulative incidence of high-dose utilization was 14.9% at 2 years after initiating hemodialysis. Among patients with ≥1 opioid prescription during follow-up, the average percentage of days exposed to high-dose utilization increased from 13.9% in 2007 to 26.1% in 2014. Compared to 0MME per day, doses < 60MME were not associated with an increased risk of mortality, but high-dose utilization was associated with a 1.63-fold (95% CI, 1.57, 1.69) increased risk of mortality. The risk of mortality associated with opioid dose was highest in the first year after hemodialysis initiation. CONCLUSIONS:The risk of mortality associated with opioid utilization among individuals on hemodialysis increases as doses exceed 60MME per day and is greatest during periods of high-dose utilization. Patients and clinicians should carefully weigh the risks and benefits of opioid doses exceeding 60MME per day.

Importance:Historically, deceased organ donation was lower among Black compared with White populations, motivating efforts to reduce racial disparities. The overarching effect of these efforts in Black and other racial/ethnic groups remains unclear. Objective:To examine changes in deceased organ donation over time. Design, Setting, and Participants:This population-based cohort study used data from January 1, 1999, through December 31, 2017, from the Scientific Registry of Transplant Recipients to quantify the number of actual deceased organ donors, and from the Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research Detailed Mortality File to quantify the number of potential donors (individuals who died under conditions consistent with organ donation). Data were analyzed from December 2, 2019, to May 14, 2020. Exposures:Race and ethnicity of deceased and potential donors. Main Outcomes and Measures:For each racial/ethnic group and year, a donation ratio was calculated as the number of actual deceased donors divided by the number of potential donors. Direct age and sex standardization was used to allow for group comparisons, and Poisson regression was used to quantify changes in donation ratio over time. Results:A total of 141 534 deceased donors and 5 268 200 potential donors were included in the analysis. Among Black individuals, the donation ratio increased 2.58-fold from 1999 to 2017 (yearly change in adjusted incidence rate ratio [aIRR], 1.05; 95% CI, 1.05-1.05; P < .001). This increase was significantly greater than the 1.60-fold increase seen in White individuals. Nevertheless, substantial racial differences remained, with Black individuals still donating at only 69% the rate of White individuals in 2017 (P < .001). Among other racial minority populations, changes were less drastic. Deceased organ donation increased 1.80-fold among American Indian/Alaska Native and 1.40-fold among Asian or Pacific Islander populations, with substantial racial differences remaining in 2017 (American Indian/Alaska Native population donation at 28% and Asian/Pacific Islander population donation at 85% the rate of the White population). Deceased organ donation differences between Hispanic/Latino and non-Hispanic/Latino populations increased over time (4% lower in 2017). Conclusions and Relevance:The findings of this cohort study suggest that differences in deceased organ donation between White and some racial minority populations have attenuated over time. The greatest gains were observed among Black individuals, who have been the primary targets of study and intervention. Despite improvements, substantial differences remain, suggesting that novel approaches are needed to understand and address relatively lower rates of deceased organ donation among all racial minorities.