Many guidelines suggest incorporating clinical assessment, imaging, and D-dimer testing into diagnostic algorithms in patients with suspected deep venous thrombosis (DVT) and pulmonary embolism (PE). This special article reviews the evidence supporting the use of algorithms and their individual components for diagnosis of upper- and lower-extremity DVT and PE in adults, including pregnant women. The authors identified evidence through several electronic database searches to April 2017, evaluated the robustness of selected evidence, assessed whether diagnostic approaches that do not use algorithms are acceptable, and identified knowledge gaps that require further research.

Background Colorectal cancer often presents with obstruction needing urgent, potentially life-saving decompression. The comparative efficacy and safety of endoluminal stenting versus emergency surgery as initial treatment for such patients is uncertain. Methods Patients with left-sided colonic obstruction and radiological features of carcinoma were randomized to endoluminal stenting using a combined endoscopic/fluoroscopic technique followed by elective surgery 1–4 weeks later, or surgical decompression with or without tumour resection. Treatment allocation was via a central randomization service using a minimization procedure stratified by curative intent, primary tumour site, and severity score (Acute Physiology And Chronic Health Evaluation). Co-primary outcome measures were duration of hospital stay and 30-day mortality. Secondary outcomes were stoma formation, stenting completion and complication rates, perioperative morbidity, 6-month survival, 3-year recurrence, resource use, adherence to chemotherapy, and quality of life. Analyses were undertaken by intention to treat. Results Between 23 April 2009 and 22 December 2014, 245 patients from 39 hospitals were randomized. Stenting was attempted in 119 of 123 allocated patients (96.7 per cent), achieving relief of obstruction in 98 of 119 (82.4 per cent). For the 89 per cent treated with curative intent, there were no significant differences in 30-day postoperative mortality (3.6 per cent (4 of 110) versus 5.6 per cent (6 of 107); P = 0.48), or duration of hospital stay (median 19 (i.q.r. 11–34) versus 18 (10–28) days; P = 0.94) between stenting followed by delayed elective surgery and emergency surgery. Among patients undergoing potentially curative treatment, stoma formation occurred less frequently in those allocated to stenting than those allocated to immediate surgery (47 of 99 (47.5 per cent) versus 72 of 106 (67.9 per cent); P = 0.003). There were no significant differences in perioperative morbidity, critical care use, quality of life, 3-year recurrence or mortality between treatment groups. Conclusion Stenting as a bridge to surgery reduces stoma formation without detrimental effects. Registration number: ISRCTN13846816 (http://www.controlled-trials.com).

No abstract