Statistics from Altmetric.com
Editor,—Birkebæck and Clausen reported auxological data in very long term survivors of acute lymphoblastic leukaemia (ALL) in first complete remission.1 We also evaluated long term survivors of childhood cancer, including acute leukaemia.2 To compare our data with those of Birkebæck and Clausen we used the same selection criteria. Twelve of 26 survivors received chemotherapy (C) alone and 14 chemotherapy plus cranial irradiation (C + I). Median age at diagnosis, at the end of treatment, and at evaluation were similar to Birkebæck and Clausen’s patients (5.3 v 4.3, 8.1v 7.25, and 19.3v 20.5 years, respectively).
Poor nutritional status at diagnosis (−1.01 body mass index standard deviation score (BMISDS) for the C group and −1.03 for the C + I group) improved during treatment (−0.23 and 0 BMISDS for each group at the end of treatment). We confirm the frequency of obesity (31% for the whole group); however, we found a tendency to higher figures in group C + I compared with group C (43% v17%), as well as a higher proportion of obese girls than boys (39%v 12.5%). The sample size precludes achieving significance. We suspect, based on our personal experience over 25 years caring for children with cancer, that parental health and illness constructs may influence the acquisition of eating habits of survivors.
At diagnosis, we also found that patients were taller than the normal population (0.42 and 0.27, mean height standard deviation score (HSDS) in the C and C + I groups, respectively). At evaluation mean HSDS were, respectively, −0.42 and −1.02 for the C and the C + I groups. Mean height loss since diagnosis was −0.84 and −1.29, respectively. In contrast, almost all survivors had achieved their target height. Accordingly, corrected mean height loss for their target height (calculated as HSDS at evaluation minus target HSDS) reduced the differences between diagnosis and evaluation to 0.17 HSDS for the C group and −0.34 HSDS for the C + I group. Height loss in the C group was observed by Sklar and colleagues,3 but not by others.1 4-6 Birkebæck and Clausen’s results would be interpreted better if they included target height in their analysis. Few studies have contrasted final height of survivors with their target height.4 7 As a result, height loss appears greater than that genetically expected so clinicians may overemphasise the adverse effect of cancer treatment (especially cranial irradiation) on survivors’ height.
Drs Birkebæck and Clausen comment:
Lopez-Andreu et al’s patient group was comparable with our group regarding age at treatment and age at follow up and they found the same frequency of obesity as we did; however, Lopez-Andreu et al did not mention changes in BMISDS from attained final height until follow up.
Mean height standard deviation score (HSDS) loss since diagnosis in the C + I group was −1.29 similar to the HSDS loss in our population of −1.27. In the C group, however, Lopez-Andreuet al found a mean HSDS loss since diagnosis of −0.84 while HSDS in our C group was the same at diagnosis and follow up. Lopez-Andreu et al did not describe the treatment regimen, which might have been more intensive than our C group regimen, explaining the discrepancy. Sklaret al using more intensive treatment than us, also found a final HSDS loss in the C group, while Holmet al, who used the same treatment regimen as us did not find any final HSDS loss in a group treated with chemotherapy only.1-1 1-2
Lopez-Andreu et al found that most of the patients attained their target height and proposed a corrected mean height loss for the target height (HSDS at final height minus target HSDS) when evaluating the impact of treatment on final height. We calculated mean height loss for target height in our population of patients treated for ALL in childhood. C group patients reached their target height, but for the C + I group mean target height was significantly higher than final height. Thus the corrected mean height loss for the target height was −0.64. Using this number the HSDS loss since diagnosis was −1.29 compared with an HSDS loss of −1.27 when not taking target HSDS into consideration. We cannot confirm the findings of Lopez-Andreu et al of a decreased target HSDS of patients treated for ALL in childhood with C + I.