Lisette van Dam

Chapter 5 84 Table 2. Clinical decision rule for upper extremity deep vein thrombosis by Constans et al Item Value Venous material* 1 point Localized pain along deep veins of the upper arm 1 point Unilateral pitting edema of the upper arm 1 point Other diagnosis at least as plausible -1 point Cut-off points Unlikely clinical probability < 2 Likely clinical probability ≥ 2 * Venous material including catheter or access device in a subclavian or jugular vein or pacemaker Statistical analysis Baseline characteristics are described as mean with standard deviation (SD) or median with interquartile range (IQR). For the primary outcome, we estimated the sensitivity of MR-NCTI for the diagnosis of UEDVT which was determined by calculating the proportion of MR scans that were read as “positive for UEDVT” in patients with CUS or (CT) venography proven UEDVT. Specificity was determined by calculating the proportion of MR scans that were read as “negative for UEDVT” in patients where UEDVT was ruled- out by either an unlikely CDR and normal D-dimer test or a normal serial CUS or normal (CT) venography both followed by a 3-month follow-up without VTE. The corresponding 95% confidence interval (95%CI) of both the sensitivity and specificity were calculated. A point estimate of the sensitivity of > 90% was defined as acceptable for initiating a future management study. We estimated that a sample size of 30 patients in each group (positive and negative UEDVT diagnosis) was needed to reach the sensitivity of greater than 90% with a corresponding 95%CI of ±15%. Therefore, we aimed to include 60 patients in total. For the secondary outcome, in which we assessed interobserver agreement of MR-NCTI reading, the κ -statistic was calculated. The kappa value for agreement was interpreted as follows: poor ( < 0.20), fair (0.21–0.40), moderate (0.41–0.60), good (0.61–0.80) or excellent (0.81–1.00).(20) Analyses were performed in SPSS version 25 (IBM, Armonk, NY, USA).