Short bowel syndrome (SBS) and enterocutaneous or enteroatmospheric fistulas are common indications for home parenteral nutrition (HPN). However, there are few data describing factors influencing surgical decision-making or outcomes particularly following fistula development. We aimed to compare outcomes between patients with SBS and fistulas and explore surgical decision-making. HPN-dependent adults from 2001–2018 at a national reference centre were included in this study. HPN cessation was analysed using death as competing risk. In total, 465 patients (SBS (62%), fistula (38%)) were included, with median HPN dependency of 2.6 years. In total, 203 patients underwent reconstructive surgery; while frailty was the commonest reason for not undergoing surgery (49.2%), 22.7% declined surgery. Overall, 170 ceased HPN, with a probability of 13.8%, 34.1% and 38.3% at 1, 5 and 10 years, respectively. Patients undergoing surgery had higher nutritional autonomy rates (109.8 incidences/1000 patient years) compared to those not undergoing surgery (18.1 incidences/1000 patient years; p < 0.001). A total of 295 patients (63.4%) were predicted to cease HPN based on gastrointestinal anatomy but only 162/295 (54.9%) achieved this; those unable to do so were older with a higher comorbidity index. There were no differences in long-term nutritional and survival outcomes or surgical decisions between patients with SBS and fistulas, or between enterocutaneous and enteroatmospheric fistulas. This study represents one of the largest datasets describing the ability of HPN-dependent patients with SBS or fistulas to achieve nutritional autonomy. While reconstructive surgery facilitates HPN cessation, approximately one-fifth of patients declined surgery despite HPN dependency. These data will better inform patient expectation and help plan alternative therapies.

Background: We aimed to identify the prevalence of acute hypoxaemic respiratory failure (AHRF) in the intensive care unit (ICU) and its associated mortality. The secondary aim was to describe ventilatory management as well as the use of rescue therapies. Methods: Multi-centre prospective study in nine hospitals in Wales, UK, over 2-month periods. All patients admitted to an ICU were screened for AHRF and followed-up until discharge from the ICU. Data were collected from patient charts on patient demographics, clinical characteristics, management and outcomes. Results: Out of 2215 critical care admissions, 886 patients received mechanical ventilation. A total of 197 patients met inclusion criteria and were recruited. Seventy (35.5%) were non-survivors. Non-survivors were significantly older, had higher SOFA scores and received more vasopressor support than survivors. Twenty-five (12.7%) patients who fulfilled the Berlin definition of acute respiratory distress syndrome (ARDS) during the ICU stay without impact on overall survival. Rescue therapies were rarely used. Analysis of ventilation showed that median Vt was 7.1 mL/kg PBW (IQR 5.9–9.1) and 21.3% of patients had optimal ventilation during their ICU stay. Conclusions: One in four mechanically ventilated patients have AHRF. Despite advances of care and better, but not optimal, utilisation of low tidal volume ventilation, mortality remains high.

Lack of expertise in home parenteral nutrition (HPN) management has been reported as a barrier to its initiation in patients with advanced cancer (AC), and there are limited data describing hospital readmissions and HPN-related complications. We aimed to assess a centralized approach for managing HPN in AC and evaluate associated outcomes, including hospital readmissions and HPN-related complications. This was a cohort study of adults with AC requiring palliative HPN between 2010–2018 at a tertiary intestinal failure (IF) center, primarily utilizing a centralized model of HPN oversight to discharge patients remotely from an oncology center to their homes over a wide geographic area. A total of 126 patients were included, with a median distance between the patient’s home and the IF center of 17.5 km (IQR 10.9–39.1; maximum 317.4 km). A total of 28 (22%) patients experienced at least one HPN-related complication, the most common being a central venous catheter (CVC) occlusion and electrolyte abnormalities. The catheter-related bloodstream infection (CRBSI) rate was 0.49/1000 catheter days. The CVC type, administration of concomitant chemotherapy via a distinct CVC lumen separate from PN, venting gastrostomy and distance between the patient’s home and the IF center were not associated with CRBSI or mechanical CVC complications. A total of 82 (65.1%) patients were readmitted while on HPN, but only 7 (8.5%) of these readmissions were HPN-related. A total of 44 (34.9%) patients died at home, 41 (32.5%) at a hospice and 41 (32.5%) in a hospital. In conclusion, this study demonstrates that a centralized approach to IF care can provide HPN to patients over a large geographical area while maintaining low HPN-related complications that are comparable to patients requiring HPN for benign conditions and low hospital readmission rates.

Recent description of the microbiology of sepsis on the wards or information on the real-life antibiotic choices used in sepsis is lacking. There is growing concern of the indiscriminate use of antibiotics and omission of microbiological investigations in the management of septic patients. We performed a secondary analysis of three annual 24-h point-prevalence studies on the general wards across all Welsh acute hospitals in years 2016–2018. Data were collected on patient demographics, as well as radiological, laboratory and microbiological data within 48-h of the study. We screened 19,453 patients over the three 24 h study periods and recruited 1252 patients who fulfilled the entry criteria. 775 (64.9%) patients were treated with intravenous antibiotics. Only in 33.65% (421/1252) of all recruited patients did healthcare providers obtain blood cultures; in 25.64% (321/1252) urine cultures; in 8.63% (108/1252) sputum cultures; in 6.79% (85/1252) wound cultures; in 15.25% (191/1252) other cultures. Out of the recruited patients, 59.1% (740/1252) fulfilled SEPSIS-3 criteria. Patients with SEPSIS-3 criteria were significantly more likely to receive antibiotics than the non-septic cohort (p < 0.0001). In a multivariable regression analysis increase in SOFA score, increased number of SIRS criteria and the use of the official sepsis screening tool were associated with antibiotic administration, however obtaining microbiology cultures was not. Our study shows that antibiotics prescription practice is not accompanied by microbiological investigations. A significant proportion of sepsis patients are still at risk of not receiving appropriate antibiotics treatment and microbiological investigations; this may be improved by a more thorough implementation of sepsis screening tools.