Background: Neuroretinal atrophy is associated with whole-brain atrophy and disease activity in multiple sclerosis (MS). Recent findings support that subclinical visual pathway involvement might also occur in neuromyelitis optica spectrum disorders (NMOSDs). Objective: The objective of this study is to assess retinal thinning in MS and NMOSD and its association with disease activity. Methods: In total, 27 NMOSD and 54 propensity-score-matched MS patients underwent optical coherence tomography, visual acuity, and visual-evoked potentials at 2.4 years apart, in addition to routine clinical and magnetic resonance imaging (MRI) assessment. We excluded eyes with acute optic neuritis. Results: In NMOSD, we detected peripapillary retinal nerve fiber layer (pRNFL) thinning in patients with disease activity during follow-up (−0.494 µm/year), but not in stable patients (−0.012 µm/year). Macular ganglion cell-inner plexiform layer (GCIPL) thinning occurred instead in all patients (−0.279 µm/year). Relapsing–remitting multiple sclerosis (RRMS) meeting NEDA-3 criteria had no pRNFL or GCIPL thinning during follow-up. Active-disease RRMS and progressive MS, both active and stable, displayed pRNFL (−0.724, −0.586, −0.556 µm/year, respectively) and GCIPL loss. Conclusion: In MS, neuroretinal atrophy was associated with disease activity but occurred in progressive MS even when achieving NEDA-3 criteria. In NMOSD, pRNFL thinning was associated with non-ocular relapses due to a spreading of inflammatory activity. GCIPL thinning was found in all patients, supporting a primary retinal pathology targeting AQP4-rich structures.

IV-PCA was superior. Total opioid reduction and removal of the infusion pumps were achieved earlier in the IV-PCA group while opioid side-effects were more frequent in the EA group. Hospital stay was shorter in the IV-PCA group (median 74 vs 104h,p<0.001) Conclusion: Overall,IV-PCA demonstrated effective postoperative analgesia non-inferior to EA with a trend towards less side-effects and significantly shorter hospital length of stay.

Introduction: Data on surgical management of breast liver metastasis are limited. We sought to determine the safety and long-term outcome of patients undergoing hepatic resection of breast cancer liver metastases (BCLM). Methods: Using a multi-institutional, international database,111 patients were identified who underwent curativeintent surgery for BCLM between 1986e2012. Clinicopathologic, operative and outcome data were collected and analyzed. Results: Median patient age was 56 years (IQR: 46e66) and most patients (93.7%) were white. Median size of the primary breast cancer was 2.5 cm (IQR: 2.0e3.0); 43 (55.1%) patients had primary tumor nodal metastasis. Among patients with metachronous disease (n = 54, 48.6%), the median time from diagnosis of the breast cancer to the resection of liver metastasis was 55.3 months (IQR: 31.2e93.5). The mean size of the largest liver lesion was 3.0 cm (2.0e4.5); most patients (54.1%) had a solitary metastasis. At surgery, almost twoethirds of patients (62.2%) underwent a minor hepatectomy; an R0 resection was achieved in most patients (91.3%). Postoperative morbidity and mortality were 23.9% and 0%, respectively. Median, 3-and 5-year disease-free survival was 24.0 months, 40.4% and 23.7%, respectively. Median, 3-and 5-year overall-survival was 53.4 months, 75.2% and 45.7%, respectively (Figure). On multivariable analysis, surgical margin status was an independent predictor of worse overall survival (HR 4.39,;P = 0.001). Conclusions: In selected patients, resection of breast cancer liver metastases can be done safely and is associated with low morbidity and zero mortality. Although recurrence occurs in about one-half of patients by 5-years following surgery, a subset of patients may derive a survival benefit, especially from a margin negative resection.

The intrahepatic biliary cystadenoma is a rare benign tumor of the liver, originating from an intrahepatic bile duct: it becomes symptomatic only when it causes obstruction of the bile duct itself. Regardless of the various diagnostic modalities available, it is difficult to distinguish preoperatively the cystadenoma both from a simple liver cyst, and from a cystic carcinoma of the bile duct. An incomplete surgical removal of the cyst often results in a higher risk of size increase and recurrence, even considering that the lesion may degenerate into a cystadenocarcinoma. Between January 2004 and May 2011, 1,173 liver resections were carried out at the Hepatobiliary Surgery Unit of San Raffaele Hospital: 12 of these were performed for cystadenoma. Forty-six patients underwent laparoscopic liver cysts deroofing: definitive histological examination in six of these patients revealed instead the diagnosis of cystadenoma. In 50% of cases, the diagnosis of cystadenoma was therefore acquired as a result of an incidental finding. The patients were all female, median age 45 years. The liver resection included six cases of left hepatectomy, three left lobectomies, and three of the right hepatectomy. The operations were performed by laparotomy, with the exception of two left lobectomies completed laparoscopically. In all cases, the postoperative course was without major complications. The resection was radical in all cases and the median hospital stay was 5 days. At a median follow-up of 16 months (range 7-30), all patients are alive and disease free. Biliary cystadenomas can easily be misunderstood and interpreted as simple hepatic cysts. Radical surgical resection is necessary and provides good short- and long-term outcomes.

In the original article Moritz Schmelzle's last name is spelled wrong. It is correct as reflected here.