Baeyens H, De Dobbeleer G, Baeyens J P
Tijdschr Gerontol Geriatr. 2001 Oct;32(5):206-9.
The purpose of this study was to assess the agreement between rectal temperature and infrared emission detection (IRED). The specificity and sensitivity, the positive predictive value and the negative predictive value were calculated. It was a prospective, observational and unblinded study. Thermoscan-pro-1 was used to collect pairs of data in 41 patients, all hospitalised on an acute geriatric ward during a 4-month period. 41% of the patients were febrile. The 'bias' between rectal and tympanic measurement (left ear) was -0.64 degree C (d) and the standard deviation 0.58 degree C (s). The '95% Confidence Interval' for the bias was -0.82 degree C and -0.46 degree C. The '95% limits of agreement' between the rectal measurement and the tympanic measurement (left ear) was -1.78 degrees C and 0.50 degree C. Regression analysis taught us that bias was increasing with increasing temperature. If we take into account one measurement with the IRED ear thermometry, the sensitivity was 41%, the specificity 83%. The positive predictive value was 64% en the negative predictive value 67%. We can conclude that the IRED ear thermometry is not a valuable alternative for the rectal, digital thermometry in the elderly, frail and febrile hospital inpatients.
本研究的目的是评估直肠温度与红外发射检测(IRED)之间的一致性。计算了特异性和敏感性、阳性预测值和阴性预测值。这是一项前瞻性、观察性且非盲法的研究。使用Thermoscan-pro-1在41例患者中收集成对数据,所有患者均在4个月期间入住急性老年病房。41%的患者发热。直肠温度与鼓膜温度(左耳)测量之间的“偏差”为-0.64摄氏度(d),标准差为0.58摄氏度(s)。偏差的“95%置信区间”为-0.82摄氏度至-0.46摄氏度。直肠测量与鼓膜测量(左耳)之间的“95%一致性界限”为-1.78摄氏度至0.50摄氏度。回归分析表明,偏差随温度升高而增大。如果考虑使用IRED耳温计进行一次测量,敏感性为41%,特异性为83%。阳性预测值为64%,阴性预测值为67%。我们可以得出结论,对于老年、体弱且发热的住院患者,IRED耳温计并非直肠指测体温的有效替代方法。