Patients triaged to neoadjuvant chemotherapy have higher rates of sarcopenia: An opportunity for prehabilitation.

OBJECTIVE: To compare baseline body composition measures (BCM), including sarcopenia, between patients with advanced epithelial ovarian cancer (EOC) undergoing primary cytoreductive surgery (PCS) versus neoadjuvant chemotherapy/interval cytoreductive surgery (NACT/ICS) and evaluate changes in BCM pre-NACT versus pre-ICS. METHODS: Patients with stage IIIC/IV EOC who underwent PCS or NACT with curative intent between 1/1/2012 and 7/31/2016 were included. Computed tomography scans were evaluated via a semi-automated program to determine BCM. Measures evaluated include skeletal muscle area (SMA), skeletal muscle density (SMD), skeletal muscle index (SMI), and skeletal muscle gauge (SMG). Sarcopenia was defined as SMI <39.0 cm2/m2. RESULTS: The study included 200 PCS patients and 85 NACT/ICS patients, of which 76 had both pre-NACT and pre-ICS scans. NACT patients were significantly more likely to be sarcopenic compared to PCS patients (40.0% vs 27.5%, p = 0.04). Mean SMA (107.3 vs 113.4 cm2, p = 0.004) and mean SMG (1344.6 vs. 1456.9 (cm2 x HU)/m2, p = 0.06) were lower in NACT patients. Among NACT/ICS patients, mean SMI significantly decreased -1.4 cm2/m2 (p = 0.005) at the time of surgery, resulting in a non-statistically significant increase in the percentage of sarcopenic patients from baseline (40.8% vs. 50.0%, p = 0.09). CONCLUSIONS: Sarcopenia is more common in patients with advanced EOC undergoing NACT compared to PCS when using an evidence-based triage system for triage decisions. Body composition changes significantly over the course of NACT. Sarcopenia may be an indicator of debility and another factor for consideration in treatment planning. Further research into body composition's effects on prognosis and altering sarcopenia is necessary.

Prediction of short-term surgical complications in women undergoing pelvic exenteration for gynecological malignancies.

OBJECTIVE: Pelvic exenteration (PE) is an extensive surgery associated with high rates of postoperative morbidity and mortality. The absence of well-defined preoperative selection criteria to identify patients eligible for PE prompted the assessment of pre-operative predictors of 30-day major surgical complications. METHODS: Demographics and surgical characteristics of patients undergoing PE for gynecologic cancer in a single institution between 01/2004-12/2016 were reviewed. Postoperative complications within 30 days following surgery were graded using the Accordion grading system. Logistic regression was used to analyze potential risk factors for severe postoperative complications. RESULTS: A total of 138 patients were included in the cohort. Forty-five patients underwent total PE, 52 anterior PE, and 41 posterior PE. Among the 137 patients with follow-up, a severe postoperative complication was experienced by 37 patients (27.0%) and 3 patients (2.2%) experienced death within 90 days. The most frequent grade 3 complications were complications of urinary reconstruction (n = 15), wound dehiscence (n = 9), and abdominal abscess requiring intervention with drain or return to the operating room (n = 6). On multivariable analysis, independent predictors of severe postoperative complications were anterior or total PE (adjusted odds ratio (aOR): 11.66, 95% CI 2.56-53.18), pre-operative hemoglobin ≤10 mg/dl (aOR 2.70, 95% CI 1.02-7.14) and presence of 3+ comorbidities (aOR: 2.76, 95% CI 1.07-7.10). CONCLUSIONS: Major complications after exenteration are common. Surgical complexity and patient selection play a considerable role in predicting complications. These data can be used to better risk stratify patients undergoing PE.

Clustering and variable selection in the presence of mixed variable types and missing data.

We consider the problem of model-based clustering in the presence of many correlated, mixed continuous, and discrete variables, some of which may have missing values. Discrete variables are treated with a latent continuous variable approach, and the Dirichlet process is used to construct a mixture model with an unknown number of components. Variable selection is also performed to identify the variables that are most influential for determining cluster membership. The work is motivated by the need to cluster patients thought to potentially have autism spectrum disorder on the basis of many cognitive and/or behavioral test scores. There are a modest number of patients (486) in the data set along with many (55) test score variables (many of which are discrete valued and/or missing). The goal of the work is to (1) cluster these patients into similar groups to help identify those with similar clinical presentation and (2) identify a sparse subset of tests that inform the clusters in order to eliminate unnecessary testing. The proposed approach compares very favorably with other methods via simulation of problems of this type. The results of the autism spectrum disorder analysis suggested 3 clusters to be most likely, while only 4 test scores had high (>0.5) posterior probability of being informative. This will result in much more efficient and informative testing. The need to cluster observations on the basis of many correlated, continuous/discrete variables with missing values is a common problem in the health sciences as well as in many other disciplines.

Sex differences in fitness and cardiac function during exercise in adolescents with chronic fatigue.

Females demonstrate less robust Frank-Starling mechanism with respect to cardiac preload than males at rest. We asked whether this phenomenon would also affect cardiac performance during exercise. We hypothesized that stroke volume (SV) response to exercise would be more limited in deconditioned females such that cardiac output would be mainly rate dependent, compared with males. We conducted a chart audit of clinical exercise tests performed by adolescents with chronic fatigue. Oxygen uptake (V˙O2) was measured breath-by-breath at rest and during cycle ergometry, while cardiac output was measured by acetylene rebreathing at rest plus 2-3 subthreshold workloads. SV response was analyzed in two ways: after normalization for body surface area (SV index, SVI) and as percentage change from resting values. Among 304 adolescents (78% females) with chronic fatigue, 189 (80%) of 236 females and 52 (76%) of 68 males were deconditioned (peakV˙O2 <90% predicted). Heart rate trajectory during exercise was steeper for unfit than fit females, 70 vs 61 beat·min-1 per L·min-1 V˙O2, (P=.003); but not for males, 47 vs 42 beat·min-1 per L·min-1 V˙O2 (P=.23). The highest measured SVI did not differ between unfit vs fit females (42.8 vs 41.5 mL·m-2 , P=.39) while fit males showed larger SV during exercise than their unfit peers (highest SVI 55.9 vs 48.0 mL·m-2 , P=.014). Both qualitative and quantitative sex differences exist in SV responses to exercise among chronically fatigued adolescents, suggesting volume loading may be more efficacious in girls.

Vaginal vs. robotic hysterectomy for patients with endometrial cancer: A comparison of outcomes and cost of care.

OBJECTIVE: To compare outcomes and cost for patients with endometrial cancer undergoing vaginal hysterectomy (VH) or robotic hysterectomy (RH), with or without lymphadenectomy (LND). METHODS: Patients undergoing planned VH (and laparoscopic LND) or RH (and robotic LND) between January 2007 and November 2012 were reviewed. Patients with stage IV disease, synchronous cancer, synchronous surgery, or treated with palliative intent were excluded. Patients were objectively triaged to LND per institutional protocol based on frozen section. Outcomes were compared between VH and RH groups matched 1:1 on propensity scores. RESULTS: VH was planned in 153 patients; 60 (39%) had concurrent LND while 93 (61%) were low risk and did not require LND. RH was planned in 398 patients; 225 (56%) required concurrent LND and 173 (44%) did not. Among 50 PS-matched pairs without LND, there was no significant difference in complications, length of stay, readmission, or progression free survival. However, median operative time was 1.3h longer and median 30-day cost $3150 higher for RH compared to VH (both p<0.001). Among patients requiring LND, 42 PS-matched pairs were identified. Median operative time was not different when pelvic and para-aortic LND was performed, and 12min longer in the VH group for pelvic LND alone (p=0.03). Median 30-day cost was $921 higher for RH compared to VH when LND was required (p=0.08). CONCLUSION: Utilization of vaginal hysterectomy for endometrial cancer results in similar surgical and oncologic outcomes and lower costs compared to RH and should be considered for appropriate patients with a low risk of requiring LND.

Abnormal sweating patterns associated with itching, burning and tingling of the skin indicate possible underlying small-fibre neuropathy.

BACKGROUND: Itching, burning, numbness and tingling of the skin are frequent reasons for dermatology consultation. We hypothesized that these sensations may be attributable to a small-fibre neuropathy. Sweating, which is mediated by small nerve fibres, may be a surrogate marker of small-fibre neuropathy. OBJECTIVES: To investigate the results of thermoregulatory sweat testing (TST), which depicts and estimates whole-body sweating, in patients with itching, burning, numbness and tingling sensations. METHODS: We retrospectively reviewed the medical records of 227 patients with itching, burning, numbness and tingling sensations involving the skin who were seen at our institution during 2008 and also underwent TST. RESULTS: The mean age of the cohort was 54 years (range 3-89), and 58% were female. In all, 149 patients (66%) had abnormal TST results; in 119 (80%) of these patients the areas of anhidrosis on TST corresponded to their symptomatic areas. For each symptom analysed separately, the area of anhidrosis correlated with the area of symptoms in most patients. CONCLUSIONS: Patients with burning, itching, numbness and tingling have abnormal sweating patterns and often do not sweat in the symptomatic areas. These novel findings suggest that a small-fibre neuropathy may underlie many cutaneous symptoms and that the neuropathy can be estimated using TST.

Risk-scoring system for the individualized prediction of lymphatic dissemination in patients with endometrioid endometrial cancer.

OBJECTIVE: To develop a risk-scoring system (RSS) for the prediction of lymphatic dissemination after hysterectomy in endometrioid endometrial carcinoma (EC). METHODS: Patients who underwent surgery from 1/1/1999-12/31/2008 were evaluated. Patients with non-endometrioid histology, stage IV with macroscopic extrauterine disease, or receiving adjuvant therapy (excluding brachytherapy) without pelvic and/or paraaortic (P/PA) lymphadenectomy (LND) were excluded. Lymph node dissemination was defined as nodal metastasis when P/PA LND was performed or P/PA lymph node recurrence after negative LND or when LND was not performed. Logistic regression analysis was used to identify predictors for lymphatic dissemination and develop a RSS and nomogram. The RSS was assessed for calibration and verified for discrimination. RESULTS: Overall, 883 patients were assessed of which 521 (59.0%) underwent P/PA LND and 57 (10.9%) had positive lymph nodes. Of patients who did not undergo P/PA LND (N=362) or had negative nodes (N=464), 10 (1.2%) patients had P/PA lymph node recurrence. Myometrial invasion, tumor diameter (TD), FIGO grade, cervical stromal invasion and lymphovascular space invasion were significant on univariable analysis. All preceding variables were included in a multivariable logistic model. A parsimonious model and an alternative full model not including TD were considered. The full model with TD (illustrated in nomogram) had the highest predictive ability (concordance index 0.88). CONCLUSION: Our RSS allows accurate quantification of the probability of lymphatic dissemination and can be used as an adjunct to clinical decision-making after hysterectomy in the absence of staging. TD is an important component of the RSS and should be routinely assessed.

Evaluating the prognostic significance of preoperative thrombocytosis in epithelial ovarian cancer.

OBJECTIVE: Preoperative thrombocytosis has been implicated as a negative prognostic marker for epithelial ovarian cancer (EOC). We assessed whether thrombocytosis is an independent risk factor for EOC recurrence and death. METHODS: Perioperative patient characteristics and process-of-care variables (National Surgical Quality Improvement Program (NSQIP)-defined) were retrospectively abstracted from 587 women who underwent EOC staging between 1/2/03-12/29/08. Thrombocytosis was defined as platelet count > 450 × 10(9)/L. Disease-free survival (DFS) and overall survival (OS) were determined using Kaplan-Meier methods. Associations were evaluated with Cox proportional hazards regression and hazard ratios (HR). RESULTS: The incidence of preoperative thrombocytosis was 22.3%. DFS was 70.8% and 36.0% at 1 and 3 years. OS was 83.3% and 54.3% at 1 and 3 years. Ascites, lower hemoglobin, advanced disease, and receipt of perioperative packed red blood cell transfusion were independently associated with thrombocytosis. Older age and the presence of coronary artery disease were associated with lower likelihood of thrombocytosis. Overall, thrombocytosis was an independent predictor of increased risk of recurrence. Among early stage (I/II) cases, there was a 5-fold increase in the risk of death and nearly 8-fold risk of disease recurrence independently associated with thrombocytosis. CONCLUSION: Preoperative thrombocytosis portends worse DFS in EOC. In early stage disease, thrombocytosis is a potent predictor of worse DFS and OS and further assessment of the impact of circulating platelet-derived factors on EOC survival is warranted. Thrombocytosis is also associated with extensive initial disease burden, measurable residual disease, and postoperative sequelae. Preoperative platelet levels may have value in primary cytoreduction counseling.

Risk factors that mitigate the role of paraaortic lymphadenectomy in uterine endometrioid cancer.

OBJECTIVE: Paraaortic lymph node (PA) dissemination in endometrial cancer (EC) is uncommon and a systematic infrarenal PA dissection carries morbidity. Our objective was to identify a subgroup of EC patients who may potentially forego PA lymphadenectomy (LND). METHODS: The study endpoint (PA Metastasis or Recurrence; PAMR) was defined as detection of metastasis to PA nodes (among those with any type of PA LND) or PA recurrence within 2 years (among patients without PA LND or those with negative nodes in the context of an inadequate (<5 nodes) PA LND). Patients with non-endometrioid histology, stage IV disease, synchronous cancers, gross extrauterine or gross adnexal disease, neoadjuvant therapy, or insufficient follow-up were excluded. Multivariable logistic regression analysis identified predictors of PAMR. RESULTS: Of the 946 patients, PAMR was observed in 4% (36/946). Multivariable analysis identified positive pelvic nodes (odds ratio (OR) 24.2; p<0.001), >50% MI (OR 5.3; p<0.001) and lymphovascular space invasion (LVSI) (OR 3.7; p=0.005) as the only three independent predictors of PAMR. When all three factors were absent (77% of study cohort), the predicted probability of PAMR was 0.6%. If intraoperative frozen section is not available on pelvic lymph nodes and LVSI, omitting PA LND in all patients with ≤ 50% MI would affect 84% (792/946) of the total cohort, with a 1.1% risk of PAMR (9/792). CONCLUSION: The majority of patients with endometrioid EC may potentially forgo PA LND with expected reductions in surgical morbidity and cost. This cohort may be identified by a combined absence of: positive pelvic nodes, >50% MI and LVSI.

Prospective assessment of survival, morbidity, and cost associated with lymphadenectomy in low-risk endometrial cancer.

OBJECTIVE: Since 1999, patients with low risk endometrial cancer (EC) as defined by the Mayo criteria have preferably not undergone lymphadenectomy (LND) at our institution. Here we prospectively assess survival, sites of recurrence, morbidity, and cost in this low risk cohort. METHODS: Cause-specific survival (CSS) was estimated using the Kaplan-Meier method and compared using the log-rank test. Complications were graded per the Accordion Classification. Thirty-day cost analyses were expressed in 2010 Medicare dollars. RESULTS: Among 1393 consecutive surgically managed cases, 385 (27.6%) met inclusion criteria, accounting for 34.1% of type I EC. There were 80 LND and 305 non-LND cases. Complications in the first 30 days were significantly more common in the LND cohort (37.5% vs. 19.3%; P<0.001). The prevalence of lymph node metastasis was 0.3% (1/385). Over a median follow-up of 5.4 years only 5 of 31 deaths were due to disease. The 5-year CSS in LND and non-LND cases was 97.3% and 99.0%, respectively (P=0.32). None of the 11 total recurrences occurred in the pelvic or para-aortic nodal areas. Median 30-day cost of care was $15,678 for LND cases compared to $11,028 for non-LND cases (P<0.001). The estimated cost per up-staged low-risk case was $327,866 to $439,990, adding an additional $1,418,189 if all 305 non-LND cases had undergone LND. CONCLUSION: Lymphadenectomy dramatically increases morbidity and cost of care without discernible benefits in low-risk EC as defined by the Mayo criteria. In these low-risk patients, hysterectomy with salpingo-oophorectomy alone is appropriate surgical management and should be standard of care.