Risk factors for infection of totally implantable central venous access ports among patients requiring port removal.

BACKGROUND: Totally implantable central venous access ports, are required for various purposes, ranging from chemotherapy to nutrition. Port infection is a common complication. In many patients with port infection, the ports are removed because antibiotics are ineffective. We evaluated the risk factors associated with port removal due to port infection. METHODS: By retrospective chart review, we collected data of 223 patients who underwent port removal for any reason. Port infection was defined as infection symptoms, such as fever; elevated white blood cell counts or C-reactive protein levels; or redness at the port site, in the absence of other infections, which improved with port removal. The characteristics of patients with or without port infection were compared using univariate (chi-squared test, t-test) and multivariate logistic regression analyses. RESULTS: We compared 172 patients without port infection to 51 patients with port infection. Univariate analysis identified sex (p = 0.01), body mass index (BMI) ⩽20 kg/m2 (p = 0.00004), diabetes mellitus (p = 0.04), and purpose of use (p = 0.0000003) as significant variables. However, male sex (p = 0.03, 95% confidence interval [CI]: 0.01-0.23), BMI ⩽20 kg m2 (p = 0.002, 95% CI: 0.06-0.29), and purpose of use (total parenteral nutrition (TPN); p = 0.000005, 95% CI: 0.31-0.76) remained significant using multivariate analysis. Moreover, the patients with short bowel syndrome and difficulty in oral intake tended to be infected easily. Additionally, Staphylococcus species were the most common microbes involved in port infection. CONCLUSIONS: Male sex, BMI ⩽20 kg/m2, and purpose of use as a TPN were risk factors for port infection. Ports should not be used for long duration of TPN or used only in exceptional cases.

Persistent Descending Mesocolon as An Intraoperative Risk Factor in Laparoscopic Surgery for Left-Sided Colon and Rectal Cancer.

INTRODUCTION: Persistent descending mesocolon (PDM) represents a failure of fusion of the descending mesentery, leading to anatomical abnormalities. This study aimed to examine the effects of anatomical features of PDM on laparoscopic surgical outcomes. METHODS: Patient backgrounds, surgical outcomes, anatomical characteristics, and operative findings were retrospectively compared between 186 patients classified into PDM and non-PDM groups who underwent primary resection for left-sided colon and rectal cancer at our hospital from January 2019 to December 2020. RESULTS: PDM was diagnosed in nine patients (4.8%). The operative time (337 ± 165 vs 239 ± 107 min, p = 0.010) was significantly different between PDM and non-PDM groups, but bleeding loss was not different (108 ± 97 ml vs 53 ± 142 ml, p = 0.259). In PDM patients, in addition to abnormal fixation of the sigmoid-descending colon junction, adhesion of the mesentery of the colon and small intestine in 100%, and adhesion between the mesocolon in 33% patients was confirmed intraoperatively. Ileus was more common in the PDM group (two cases, 22%) and in the non-PDM group (10 cases, 5.6%), but there was no significant difference in overall postoperative complications between the two groups (p = 0.215). The duration of postoperative hospital stay (28 ± 20 vs 16 ± 11 days, p = 0.002) was significant between the two groups. The left colonic artery (LCA) could not be preserved in six patients in the PDM group, one of whom had anastomotic leakage and two of whom required additional resections due to intraoperative intestinal blood flow failure. CONCLUSION: PDM prolonged operative times and duration of postoperative stay in laparoscopic surgery for left-sided colon and rectal cancer. Division of the LCA in PDM patients should be considered an intraoperative risk factor for injury to the marginal artery.

Numerical study on adjusting parameters to improve gaze estimation using planar approximations from electro-oculogram signal voltage ratios.

Gaze or eye movements are used as a communication interface in daily life. Herein, we developed a simple method for gaze estimations based on planar approximations of voltage ratios calculated from multiple electro-oculogram signals not affected by drift phenomena, which decrease accuracy. Subsequently, we conducted simulations using an eyeball battery model and investigated the effects of adjusting electrode arrangements, determination coefficients for planar approximations of voltage ratios, and threshold values for angles between simultaneous linear equations, to improve the estimation accuracy and decreased the number of required electrodes. Numerical experiments were used to identify arrangements of six electrodes with errors that were approximately 5° less than those of nine-electrode L-shaped arrangements, indicating improved estimation accuracy with fewer electrodes.