Coyne Christopher J, Le Vivian, Brennan Jesse J, Castillo Edward M, Shatsky Rebecca A, Ferran Karen, Brodine Stephanie, Vilke Gary M
Department of Emergency Medicine, University of California San Diego School of Medicine, San Diego, CA.
Department of Emergency Medicine, University of California San Diego School of Medicine, San Diego, CA.
Ann Emerg Med. 2017 Jun;69(6):755-764. doi: 10.1016/j.annemergmed.2016.11.007. Epub 2016 Dec 29.
Although validated risk-stratification tools have been used to send low-risk febrile neutropenic patients home from clinic and inpatient settings, there is a dearth of research evaluating these scores in the emergency department (ED). We compare the predictive accuracy of the Multinational Association for Supportive Care in Cancer (MASCC) and Clinical Index of Stable Febrile Neutropenia (CISNE) scores for patients with chemotherapy-induced febrile neutropenia and presenting to the ED.
We conducted a retrospective cohort study to evaluate all patients with febrile neutropenia (temperature ≥38°C [100.4°F], absolute neutrophil count <1,000 cells/μL) who presented to 2 academic EDs from June 2012 through January 2015. MASCC and CISNE scores were calculated for all subjects, and each visit was evaluated for several outcome variables, including inpatient length of stay, upgrade in level of care, clinical deterioration, positive blood culture results, and death. Descriptive statistics are reported and continuous variables were analyzed with Wilcoxon rank sum.
During our study period, 230 patients presented with chemotherapy-induced febrile neutropenia. The CISNE score identified 53 (23%) of these patients as low risk and was highly specific in the identification of a low-risk cohort for all outcome variables (98.3% specific, 95% confidence interval [CI] 89.7% to 99.9%; positive predictive value 98.1%, 95% CI 88.6% to 99.9%). Median length of stay was shorter for low-risk versus high-risk CISNE patients (3-day difference; P<.001). The MASCC score was much less specific (54.2%; 95% CI 40.8% to 67.1%) in the identification of a low-risk cohort.
Our results suggest that the CISNE score may be the most appropriate febrile neutropenia risk-stratification tool for use in the ED.
尽管已使用经过验证的风险分层工具将低风险发热性中性粒细胞减少患者从门诊和住院环境送回家，但在急诊科（ED）评估这些评分的研究却很匮乏。我们比较了多国癌症支持治疗协会（MASCC）评分和稳定型发热性中性粒细胞减少临床指数（CISNE）评分对化疗引起的发热性中性粒细胞减少且前往急诊科就诊患者的预测准确性。
我们进行了一项回顾性队列研究，以评估2012年6月至2015年1月期间前往两家学术性急诊科就诊的所有发热性中性粒细胞减少患者（体温≥38°C [100.4°F]，绝对中性粒细胞计数<1000个细胞/μL）。计算所有受试者的MASCC和CISNE评分，并对每次就诊评估几个结局变量，包括住院时间、护理级别升级、临床恶化、血培养结果阳性和死亡。报告描述性统计数据，连续变量采用Wilcoxon秩和检验进行分析。
在我们的研究期间，230例患者出现化疗引起的发热性中性粒细胞减少。CISNE评分将其中53例（23%）患者识别为低风险，并且在识别所有结局变量的低风险队列方面具有高度特异性（特异性98.3%，95%置信区间[CI] 89.7%至99.9%；阳性预测值98.1%，95% CI 88.6%至99.9%）。低风险CISNE患者的中位住院时间比高风险患者短（相差3天；P<.001）。MASCC评分在识别低风险队列方面的特异性要低得多（54.2%；95% CI 40.8%至67.1%）。
我们的结果表明，CISNE评分可能是急诊科中最适合用于发热性中性粒细胞减少的风险分层工具。
