This study aimed to objectively assess the association between venous Hb values and skin/mucosal color. Consequently, the color of the lower lip, lower palpebral conjunctiva, palm, nail bed, and inside of the upper arm demonstrated an association with Hb values. However, the results suggest that not all sites are equally crucial for evaluating anemia.
We examined the association between venous Hb values and individual color elements, categorizing them into L*, a*, and b*. However, the relevance was small except for the a* value. In the lower lip, lower palpebral conjunctiva, palm, and inside of the upper arm, the a* value, an indicator of redness, exhibited a positive association with venous Hb values. However, when comparing the normal group and the anemic group, only the lower lip and lower palpebral conjunctiva showed a significant difference in the a* value. Two potential reasons are considered: (1) The a* value of the palm and inside of the upper arm may have a weak association with venous Hb values, possibly resulting in difficulty expressing the colors of these sites as red intensity even in the presence of anemia, and (2) the small number of participants with anemia might lead to inaccurate color representation at lower Hb values.
In the nail and palm, the b* value, an indicator of yellowness, displayed a negative association with venous Hb values. However, when comparing the normal group and the anemic group, only the palm showed a significant difference in the b* value. At these sites, a lower venous Hb value reduced the red color, emphasizing the yellowness of the skin's original color. Regarding the L* value, an indicator of lightness, no significant difference was observed at any site between the normal and anemic groups, although in the palm and inside of the upper arm, the L* value showed a negative association with venous Hb values. It is suggested that Hb, being red in color, is slightly associated with the L* value.
For easier comprehension, the color of each part was converted into color difference (ΔE*), a composite value of Lab*, and analyzed within each Hb value range. If the color difference between the group with Hb < 11.9 and the group with 12.0 or more is 6.0 or more, it reaches a level where the difference in color is clearly noticeable to the human eye. Therefore, in the physical examination of anemia, it was determined that the palpebral conjunctiva and the lower lip are the most suitable sites. Additionally, for anemia diagnosis, pallor in the nail, hand, and inside of the upper arm may be discerned by trained human eyes.
When comparing between sexes, in women, the mean Hb value was lower, the a* value of the lip was lower, and the b* values of the nail and palm were higher. However, in the comparison between the normal group and the anemic group, as there was no difference in the b* value of the nail, it may indicate a distinction between sexes rather than an association with venous Hb values. The result that women had a higher L* value and lower b* value in the inside of the upper arm is also challenging to interpret. Since there was no significant difference in color in the inside of the upper arm between the normal group and the anemia group, there is a possibility that a specific skin color characteristic is only evident in Japanese women. However, formulating this conclusion is difficult due to the lack of previous studies that colorimetrically examine Japanese skin color.
According to the findings of this study, the lower palpebral conjunctiva and the lower lip, traditionally used in physical examinations, are also suitable for colorimetric evaluation in diagnosing anemia. In the nail, palm, and inside of the upper arm, anemia may be detected by trained human eyes if the venous Hb value is less than 10.0. Therefore, when utilizing these sites for anemia diagnosis, oversight is possible.
Strengths and limitations of this study
The present study also had several strengths. First, this is the first study that investigated the association between the color of the lower lip, lower palpebral conjunctiva, palm, nail bed, and inside of the upper arm measured using spectrophotometer and venous Hb values among adult Japanese. Second, we provided valuable information to primary care physicians by identifying the most appropriate site for assessing anemia based on principles of color science. Third, by presenting the association between color and Hb values as objective numerical values, we provided the foundational data necessary for future research on non-invasive Hb value prediction.
However, the present study had several limitations. First, the study included a small sample size of patients with moderate to severe anemia (Hb < 10.0 g/dL). Thus, the usefulness of the nail bed, palm, and upper arm pallor in detecting anemia cannot be entirely discounted. Second, most of the participants were elderly, and their age distribution was biased. Third, since all participants were Japanese, there is bias in ethnicity. The interpretation of the results for skin color can only be applied to the Japanese. Fourth, this study assessed pallor using a spectrophotometer, which raises the question of whether the human eye can detect the same subtle differences in pallor color. Consequently, further research is necessary to determine whether human observers can reliably identify the presence or absence of anemia through training.