To the Editor: We read the paper by Campbell-Thompson et al with great interest [1]. In this study, they used relative pancreas weight (RPW; pancreas weight [g]/body weight [kg]), attempting to normalise the effects of age, ethnicity and body weight. As a result, RPW showed no association with age and was significantly reduced in patients with type 1 diabetes. There was no correlation between RPW and disease duration.

We have previously reported a comprehensive picture of the physiological and pathological changes in pancreas volume (PV) based on computed tomography (CT) scan images of >2,000 participants [2–4]. PV was ∼10 cm3 in infancy and linearly increased with age in childhood, presumably along with somatic growth. PV then reached a plateau at ∼70 cm3 in adulthood (assuming 1 cm3 = ∼1 g). In adults, PV was positively correlated with BMI. Therefore, it appears reasonable to use RPW to adjust for the effects of age and BMI.

However, it should be noted that PV then started to decline with age from ∼60 years of age. Therefore, caution is needed when RPW is applied to older individuals; in Fig. 1a of their paper, Campbell-Thompson et al presented the relationship between RPW and age only in individuals of age ≤60 years [1].

It should also be noted that using CT scan images we measured parenchymal volume and fat volume of the pancreas separately [2]. We could determine from this that parenchymal and fat components increased proportionally in childhood, while the fat/parenchymal ratio increased with age in adulthood. In adults, the fat/parenchymal ratio also increased with obesity, suggesting that in adults, the fat rather than the parenchymal component is mainly responsible for any increase in PV with age and body weight gain. Therefore, it is important to consider the possibility that the components of the pancreas differ between children and adults even when RPW is the same.

Finally, in the Campbell-Thompson paper, RPW in those with type 2 diabetes was not significantly different from that in nondiabetic controls [1]. We have reported that total PV and parenchymal PV, but not fat volume, were significantly reduced in individuals with type 2 diabetes compared with age-, sex- and BMI-matched controls [2]. This inconsistency between studies may be due to the smaller sample size in the study by Campbell-Thompson et al (n = 40 vs 165 with type 2 diabetes). However, in their study there was also a large difference in BMI between people with type 2 diabetes and controls (33.0 vs 22.5 kg/m2, respectively) [1], suggesting that increased fat in the pancreas due to obesity is likely to have masked the reduced parenchymal pancreas weight in those with type 2 diabetes. Since the pancreas is infiltrated with adipose tissue (i.e. inter- and intra-lobular fat) [2, 5], it is difficult to completely remove fat from the resected pancreas.

Nonetheless, an attempt to develop a measure of normalised pancreas mass is of great interest, and we agree that RPW is simple to calculate and will be a new way to compare pancreas mass among groups of different age, ethnicity and body weight. The absence of an apparent correlation between RPW and disease duration in both type 1 and type 2 diabetes may reflect the fact that beta cell function/mass is already reduced by 80–90% at the time of clinical onset [6, 7], which is obviously an area of active investigation.