Developmental Cognitive Neuroscience Program, Department of Psychiatry, University of British Columbia, Vancouver, British Columbia, Canada.
Sentinel Secondary, West Vancouver, British Columbia, Canada.
PLoS One. 2021 Feb 3;16(2):e0245257. doi: 10.1371/journal.pone.0245257. eCollection 2021.
Despite the increasing personalization of medicine, surprisingly ~37.0°C (98.6°F) continues as the estimate of normal temperature. We investigated between-subject and within-subject thermal variability, whether a significant percentage of individuals have a low mean oral temperature, and whether these differ by sex, age, time of day, ethnicity, body mass index (BMI), or menstrual phase. Oral temperature was measured by Life Brand® Fast-Read Digital Oral Thermometers and sampled 14 times over two weeks, seven morning and seven evening readings. The volunteer sample consisted of 96 adults (42 men, 54 women; 27 couples, 42 singletons), ages 18-67 years. We found sizeable individual differences in body temperature and that the normal temperature of many individuals is considerably lower than 37.0°C (98.6°F). Mean temperatures ranged from 35.2°C (95.4°F) to 37.4°C (99.3°F). The mean temperature across all participants was 36.1°C (97.0°F)-lower than most studies have reported, consistent with recent evidence of temperature declining over decades. 77% had mean temperatures at least 0.55°C (1°F) lower than 37.0°C (98.6°F). Mean temperature did not differ by age, but women had higher temperatures than men, even within a couple with room temperature and warmth of clothing equated. Although oral temperature varied widely across individuals, it showed marked stability within individuals over days. Variability of temperature over days did not differ by sex, but was larger among younger adults. Using 37.0°C (98.6°F) as the assumed normal temperature for everyone can result in healthcare professionals failing to detect a serious fever in individuals with a low normal temperature or obtaining false negatives for those individuals when using temperature to screen for COVID-19, mistaking their elevated temperature as normal. Some have called for lowering the estimate of normal temperature slightly (e.g., 0.2°C [0.36°F]). That still seems an overly high estimate. More important, using any standardized "normal" temperature will lead to errors for many people. Individual differences are simply too great. Personalizing body temperature is needed. Temperature could be measured at yearly doctor visits, as blood pressure is now. That would be simple to implement. Since our results show marked thermal stability within an individual, sampling temperature only once yearly could provide an accurate indication of a person's normal temperature at that time of day. Such records over time would also provide a more accurate understanding of how temperature changes over the lifespan.
尽管医学的个性化程度不断提高，但令人惊讶的是，~37.0°C（98.6°F）仍然是正常体温的估计值。我们研究了个体间和个体内的体温变异性，是否有相当比例的个体的口腔平均温度较低，以及这些因素是否因性别、年龄、一天中的时间、种族、体重指数（BMI）或月经周期而异。口腔温度由 Life Brand®快速读数数字口腔体温计测量，并在两周内 14 次采样，每天早晚各 7 次。志愿者样本由 96 名成年人（42 名男性，54 名女性；27 对夫妇，42 名单身）组成，年龄在 18-67 岁之间。我们发现体温存在相当大的个体差异，许多人的正常体温明显低于 37.0°C（98.6°F）。平均温度范围在 35.2°C（95.4°F）至 37.4°C（99.3°F）之间。所有参与者的平均温度为 36.1°C（97.0°F），低于大多数研究报告的温度，与近年来体温下降的证据一致。77%的人的平均体温至少比 37.0°C（98.6°F）低 0.55°C（1°F）。平均体温不因年龄而异，但女性的体温高于男性，即使在室温下，穿着相同的衣服，情况也是如此。尽管个体间的口腔温度差异很大，但个体在几天内的体温变化具有明显的稳定性。个体间的体温变异性不因性别而异，但在年轻成年人中更大。如果将 37.0°C（98.6°F）作为每个人的假定正常体温，那么医护人员可能无法检测到体温较低的个体的严重发热，或者在使用体温筛查 COVID-19 时，这些个体的假阴性结果会被误认为是正常的体温升高。有人呼吁将正常体温的估计值略微降低（例如，降低 0.2°C [0.36°F]）。这似乎仍然是一个过高的估计值。更重要的是，使用任何标准化的“正常”体温都会导致许多人的错误。个体差异实在太大了。需要根据个人情况来调整体温。体温可以像现在测量血压一样，每年在医生就诊时测量一次。这将很容易实施。由于我们的研究结果表明个体内部的体温稳定性显著，因此每年仅采样一次体温就能准确反映个体在当时的正常体温。随着时间的推移，这些记录将更准确地了解体温在整个生命周期中的变化。
