Table 6.
Results of studies that assessed the efficacy or effectiveness of growth hormone treatment on bone in adults with Prader-Willi syndrome
| Author | Outcome parameter | Result | Limitations/remarks | |
|---|---|---|---|---|
| Bone mineral density | ||||
| RCTs | Jørgensen et al (2013) (44) | Change in BMDTS SDS (% ± SD); change in BMDLS SDS, % ± SD; change in BMDhip SDS, % ± SD | 12 mo of GH treatment resulted in a decrease of –1.0 ± 3.2% in BMDTS and –0.4 ± 8.0% in BMDhip. Placebo for 12 mo resulted in a decrease of –0.2 ± 2.1% in BMDTS and –1.8 ± 4.8% in BMDhip. Differences between GH treatment and placebo were nonsignificant (P > .05 for both). For BMDLS, there was a significant difference between 12 mo of GH treatment and placebo (–2.1 ± 3.4% and 1.9 ± 3.4%, respectively; P < .01). Continuous GH treatment for 24 mo did not result in any changes in BMDTS, BMDLS, or BMDhip (P > .05 for all). | Participants who dropped out were excluded from analyses. |
| Donze et al (2018)a (61) | BMDTB SDS, BMADLS SDS, BMDTB, g/cm2, BMADLS, g/cm2 | No significant difference in BMDTB SDS and BMADLS SDS between 12 mo of GH treatment and placebo (P = .51 and P = .37, respectively). BMDTB and BMADLS also did not significantly differ between 12 mo of GH treatment and placebo (data not shown). Independent of GH treatment or placebo, BMDTB SDS did not change during study (24 mo; P = .20), but BMADLS SDS decreased significantly compared to baseline (P < .01). | All participants were on GH treatment at time of inclusion. | |
| NRCTs | Butler et al (2013) (52) | BMDTB, g/cm2; mean ± SD | During 12 mo of GH treatment, BMDTB decreased from 1.14 ± 0.05 to 1.12 ± 0.05 (P > .05). | No information on missing data. |
| Khare et al (2014) (62) | BMDLS SDS | Individuals > 15 y who were treated with GH had a higher BMDLS SDS than individuals > 15 y who were never treated with GH (P = .02). | Children and adults were included. Unclear whether “on GH treatment” group also contains participants who received only GH in the past. | |
| Longhi et al (2015) (54) | BMC, g; BMDTB, g/cm2; BMDLS, g/cm2; BMDFN, g/cm2 | Individuals who were treated with GH (either in the past or at time of assessment) had a higher BMC than individuals never treated with GH (2184 ± 278 g and 1743 ± 601 g, respectively; P = .003). No differences between GH-treated and GH-untreated individuals for BMDTB (1.11 ± 0.09 and 1.15 ± 0.12, respectively; P > .05), BMDLS (1.02 ± 0.12 and 1.05 ± 0.16, respectively; P > .05), and BMDFN (0.86 ± 0.13 and 0.91 ± 0.16 respectively, P > .05). | GH treatment could either be current or past (during childhood). No information on missing data. | |
| Markers of bone formation, bone resorption, and bone regulatory markers | ||||
| RCTs | Jørgensen et al (2013) (44) | PINP, µg/L; osteocalcin, µg/L; NTX, nM | 12 mo of GH treatment resulted in a significant increase of PINP and osteocalcin (P < .001 and P < .05, respectively). NTX did not change during GH treatment (P > .05). | Participants who dropped out were excluded from analyses. |
| NRCTs | Brunetti et al (2018) (63) | RANKL, pg/mL; OPG, pg/mL; sclerostin, pg/mL | Multiple regression analysis showed GH treatment results in decreased levels of RANKL (ρ = –0.551; P < .0001) and increased levels of OPG (ρ = 0.392; P < .0001) when adjusted for age. No correlation between sclerostin and GH treatment (P > .05). | No information on participant selection. |
| Bone geometry and bone strength | ||||
| NRCTs | Longhi et al (2015) (54) | Outer diameter, mm; inner diameter, mm; MI; CSA, mm2; CA, mm2; BBRI, mm3. All parameters were evaluated at level of second metacarpal bone at its narrowest site. | Individuals who were treated with GH (either past or at time of assessment) had, compared to GH-untreated individuals, a bigger outer diameter (7.8 ± 0.7 and 7.2 ± 0.5, respectively; P = .004) and inner diameter (3.8 ± 0.8 and 3.3 ± 0.6, respectively; P = .000), and consequently a higher CSA (49.2 ± 8.9 and 40.4 ± 6.6, respectively; P = .005), CA (36.9 ± 6.5 and 31.9 ± 4.9, respectively; P = .02), MA (12.3 ± 4.8 and 8.9 ± 3.1; P = .03), and BBRI (467.3 ± 123.6 and 357.3 ± 74.2, respectively; P = .008). No difference in MI between GH-treated and GH-untreated individuals (0.6 ± 0.1 and 0.5 ± 0.1, respectively; P > .05). | GH treatment could either be current or past (during childhood). No information on missing data. |
Abbreviations: BBRI, bending breaking resistance index (bone strength); CA, cortical area; CSA, total cross-sectional area; BMADLS, lumbar spine bone mineral apparent density; BMC, bone mineral content; BMDFN, femoral neck bone mineral density; BMDhip, total hip bone mineral density; BMDLS, lumbar spine bone mineral density; BMDTS, total body bone mineral density; GH, growth hormone; MA, medullary endocortical area; MI, metacarpal index; NRCTs, nonrandomized (un)controlled studies; NTX, cross-linked n-telopeptides of type I collagen; OPG; osteoprotegerin; PINP, N-terminal propeptide of type I procollagen; RANKL, receptor activator of nuclear factor-κB ligand; RCTs, randomized clinical trials; SDS, SD scores (Z score).
a This study compares continuation of GH treatment with cessation of GH treatment, and restart of GH treatment with cessation of GH treatment.
b Bone formation markers: PINP, osteocalcin. Bone resorption marker: NTX. Bone regulatory markers: RANKL, OPG, sclerostin.