SEARCH

SEARCH BY CITATION

Keywords:

  • attitudes;
  • final height;
  • tall stature;
  • temporal changes;
  • treatment

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Aim:  Extreme tall stature may lead to a variety of concerns in tall children and their parents, leading to requests for treatment to reduce final height in some children. This study reviews referrals for tall stature to a single pediatric endocrinologist and results of treatment over 25 years from 1980 to 2004.

Methods:  Diagnoses, heights, target heights and estimated final heights at presentation, and final heights and complications in treated patients, were examined by retrospective chart review.

Results:  Of 345 referrals, 244 (71%) were girls and 101 (29%) were boys. Of the 68 (19.7%) treated, 53 (78%) were girls and 15 (22%) were boys. Most children had familial tall stature. Treated children were tall for their already tall families. Treatment reduced final heights compared with estimated final heights by (mean +/− standard error of the mean) 4.2 +/− 0.5 cm (P = 0.001) in girls and 5.1 +/− 0.8 cm (P < 0.001) in boys. Minor complications occurred in 27 (51%) girls, including 5 (9.4%) who stopped treatment because of weight gain, and 5 (33%) of boys. In more recent years, girls (but not boys) presented less frequently, were taller at presentation and opted for treatment less often, and at taller estimated final heights than in the earlier years.

Conclusions:  Any benefits of high-dose sex steroid treatment of tall children in terms of reduced final height and improved self-image are at the expense of complications in many. Fewer tall girls being referred and treated probably reflects altered attitudes to tallness in society. Such treatment should seldom – if ever – be used in the future.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Concerns reported by tall children and their parents have included social unattractiveness, feeling different, being teased, withdrawing socially, and perceived or predicted difficulties in finding partners and in purchasing clothes. Women who were as tall as adolescents have a higher prevalence of major depression than the general population with self-reported difficulties during adolescence being a factor.1 Such concerns have led to attempts to limit final height (FH) by high-dose sex steroid treatment.

Of the many reports of treatment for tall stature since the first in 1956,2 few have documented the experience of a single clinician. Such individual experience allows for consistency in clinical approach to the problem and in the advice given to the family and the patient. This consistency ensures that any changes in treatment patterns over time are not because of conflicting advice from different clinic members. Reported here are the results of 345 referrals for tall stature to the private paediatric endocrinology practice of the author in Brisbane, Australia over 25 years, and the results of treatment of 68 (19.5%) of the children.

Key Points

  • 1
    High-dose sex steroid treatment significantly reduces final height in tall children.
  • 2
    Changes over the 25 years of this study in referral numbers, heights at which referrals were made, and estimated final heights on which the families opted for treatment indicate profound alterations in attitudes to tall stature over that interval, particularly in girls.
  • 3
    Any benefits in terms of reduced final height and improved self-image are at the expense of short- and long-term complications in many. Such treatment should seldom, if ever, be used in the future.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Between January 1980 and December 2004, the author conducted paediatric endocrinology clinics in the two teaching children's hospitals in Brisbane, Australia and in a private practice, geographically separate from the hospitals. Only patients referred by their family physicians or by other pediatricians were seen. From 1996–1997, other paediatric endocrinologists also attended the hospital clinics, potentially leading to the inconsistency of clinical approach and advice to families and patients. Therefore, only patients referred to the private practice are reported here.

All interviews, measurements, examinations, bone-age readings and estimations of FH (EFH; by the Bayley–Pinneau method), diagnoses and discussions were performed by the author. Consistent information was given to families and patients regarding the diagnosis, limitations of the EFH, details of treatment protocols, and potential advantages and disadvantages of high-dose sex steroid treatment for non-disease (i.e. tall stature) based on information available at the time. Frequently, this was done on several visits. They were then advised to make a considered decision on whether to treat or not, and to convey that decision to the author at a later date. If the family opted for intervention, all treatments were prescribed and supervised by the author. Children not being treated were not followed up.

Treated children had a bone-age and EFH estimation repeated every 6 months. Girls were treated with oral ethinyl oestradiol 200 µg per day, increasing to 300 µg per day if six monthly EFH estimations were not reducing. From 1998, when ethinyl oestradiol was taken off the market, conjugated equine oestrogen was prescribed at doses of 1.25–3.75 mg. Norethisterone 2.5 or 5 mg per day for 10 days a month was also prescribed. Boys were treated with intramuscular (IM) testosterone esters 250 mg every 3 weeks, increasing to 250 mg every 2 weeks if six monthly EFH estimations were not reducing.

Patients were asked to return after 3 months until FH was reached (defined as no gain in height over 9 months or a gain in height of <1 cm over 12 months) or until the family decided to cease treatment. All patients were asked to return in a further 12 months. Not all families adhered to this advice for various reasons and FH data was not available on those children.

National Center for Health Statistics (NCHS) standards were used for height, weight and body mass index (BMI).3 Data on referrals were obtained from a patient database maintained by the author. Data on treated patients were obtained from patient files. Pre- and post-treatment data were analysed by unpaired or paired t-test and trends across time periods by analysis of variance. Statistical significance was defined as a P-value of <0.05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Patients referred

Over the 25 years, 345 children were referred for tall stature, including 244 (70%) girls and 101 (30%) boys. Diagnoses made were familial tall stature (FTS) (85%), Marfan syndrome (4%), constitutional advance in growth (3%), precocious puberty (2%), and Sotos syndrome, Klinefelter syndrome and non-classical congenital adrenal hyperplasia (1% each).

Patients treated

Families opted for treatment in 68 of the 348 patients (19.5%). Of those treated, 53 (78%) were girls and 15 (22%) boys. The percentage of children referred who were eventually treated was similar in girls (22%) and boys (15%). Marfan syndrome was present in 5 girls and 1 boy, and Klinefelter syndrome in 1 boy. The rest of the treated children (90%) had FTS. Final height data was available in 37 (70%) girls and 11 (73%) boys. There was no difference in age, height standard deviation score (SDS) and EFH between those with FH data and those without, both for girls and boys; thus, pretreatment data for the entire group is combined.

Details of pretreatment age, height, body mass index (BMI) and EFH, target height (TH), duration of treatment and last available BMI in all patients and FH, where available, are shown in Table 1. Girls and boys presented with a similar height (mean +/− standard error of the mean) SDS (2.7 +/− 0.1 and 2.7 +/− 0.2, respectively) but girls opted to start treatment with a lower mean EFH SDS (3.1 +/− 0.7) than boys (3.6 +/− 0.9) (P = 0.03).

Table 1.  Patient characteristics and duration and results of treatment
 GirlsSEM (Range)BoysSEM (Range)
  1. Pre and post-treatment comparisons: *P < 0.001;**P < 0.001;^ and ^^ not significant. Comparisons between girls and boys:***P = 0.03. Age, pre-treatment height SDS and EFH were similar in FH and no FH groups in both girls and boys so all were included in pre-treatment data.

  2. BMI, body mass index; EFH, estimated final height; FH, final height; SDS, standard deviation score; SEM, standard error of the mean.

Pre-treatmentn = 53 n = 15 
 Age (year)12.50.4 (8.7–15.4)14.20.5 (11.1–17.3)
 Height SDS2.70.1 (1.0–4.7)2.70.2 (1.6–3.5)
 BMI SDS0.1^0.2 (−1.8–4.4)−0.3^^0.2 (−1.8–1.5)
 EFH (cm)182.5*0.6 (174.0–190.3)200.5**1.6 (192.8–212.1)
 EFH SDS3.1***0.7 (1.4–4.4)3.6***0.7 (2.4–5.3)
 Target height SDS1.50.1 (0–2.9)1.60.2 (0.7–2.8)
Post-treatmentn = 37 n = 11 
 Treatment duration (mo)211.3 (4–36)18.53.0 (6–37)
 FH (cm)178.2*0.7 (170.0–187.0)195.4**1.4 (188.0–202.0)
 EFH–FH (cm)4.20.5 (−2.3–13.4)5.1^^0.8 (2.0–9.4)
 BMI SDS−0.2^0.2 (−1.8–1.9)0.10.2 (−1.5–1.2)

As shown in Figure 1 and Table 1, the treated children were from tall families (EFH SDS range 0–2.9 in girls and 0.70–2.8 in boys) but the children were taller, even allowing for their parents' heights (height SDS range 1.0–4.7 in girls and 1.6–3.5 in boys; P < 0.001).

image

Figure 1. Comparison of presenting height and target height in treated girls and boys. Mean presenting height standard deviation score (SDS) was significantly taller than mean target height SDS (P < 0.001) indicating that the treated children were tall for their already tall families.

Download figure to PowerPoint

The mean apparent reductions in FH of 4.2 +/− 0.5 cm in girls and 5.1 +/− 0.8 cm in boys were both significant (P < 0.001).

Girls without FH data had a higher (P < 0.01) pretreatment BMI SDS and post-treatment BMI SDS (0.7 +/− 0.3 and 1.1 +/− 0.3, respectively) than girls with FH data (−0.3 +/− 0.2 and −0.2 +/− 0.2, respectively). No such BMI SDS differences were evident between boys with and without FH data.

Temporal trends

Temporal trends in gender differences are shown in Figures 2–4 and Table 2. Figure 2 shows that referrals declined in later years so that referrals for 2000–2004 were only 28% of those in 1985–1989. This fall in referrals over time was more marked in girls than boys (P = 0.001). Figure 3 shows a similar rise in treated cases initially, followed by an even greater decline over recent years. No girls have started treatment since 1999 and no boys since 2002. The falls over time in both the percentage of referrals who were girls and the percentage of girls who were treated are significant (P = 0.001 in each), as shown in Table 2. Such changes over time in boys were not significant. In girls, the mean height SDS at presentation rose (P = 0.02) and the mean EFH SDS also rose (P < 0.001) over the study period (Fig. 4).

image

Figure 2. Referrals for tall stature over time. Total = 345; girls = 244 (71); Boys = 101 (29%).

Download figure to PowerPoint

image

Figure 3. Children referred and children treated divided into 5-year intervals. The falls over time in both referrals and children treated was significant in girls (P = 0.001) but not boys.

Download figure to PowerPoint

image

Figure 4. Changes over time in girls' pretreatment and estimated final height SDS divided into 5-year intervals. Changes in both were significant (P = 0.02 and <0.001, respectively).

Download figure to PowerPoint

Table 2.  Temporal trends in referrals, patients treated, pretreatment height SDS and estimated final heights
 80–8485–8990–9495–9900–04 
  1. EFH, estimated final height; ns, non-significant; SD, standard deviation; SDS, standard deviation score.

Referrals      
 Total50114876532 
 Girls3991554415 
 Boys1123322114 
 % Girls7880636846 
 % Boys2220373244 
Treated      
 Total18241691 
 Girls13191190 
 Boys35511 
 % Girls382120180 
 % Boys27221657 
Height SDS (mean (SD))      
 Girls2.3 (0.9)2.5 (0.7)3.1 (0.7)3.1 (0.5) P = 0.02
 Boys2.1 (0.5)2.8 (0.8)2.6 (0.8)  ns
EFH cm (mean (SD))      
 Girls180.0 (3.7)181.8 (4.1)183.6 (4.8)186.9 (1.9) P = 0.001
 Boys199.3 (6.6)199.7 (6.9)201.9 (7.7)  ns

Thus, over time, fewer girls were seen, the mean presenting height of girls was taller, girls were less likely to want treatment and the EFH which led to treatment was taller – differences were not seen in boys.

Complications

Minor complications during treatment were reported in 27 girls (51%; often headaches, weight gain, nausea, dysmenorrhoea and hirsutism) and 5 boys (33%; acne and oedema). However, the only complication leading to cessation of treatment was unacceptable weight gain in five girls (9.4%) who ceased treatment after 6–13 months.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The report details the experience of one clinician over 25 years in seeing 345 and treating 68 children with tall stature. The author was personally responsible for all aspects of the clinical experience of the children, with a consistent approach to the problem throughout. This involved providing as much information as possible to the families and patients and insisting that they make the decision to treat or not to treat themselves.

Eighty-five percent of the referred children and 90% of those treated had FTS. The treated children were from tall families and had EFHs >97 percentile. Of interest is that they were unusually tall even for their tall parents. Reasons for this are not clear but studies looking at whether they have inherited multiple copies of genes coding for tallness would be interesting.

Unfortunately, FH data was not available in 39% of girls and 27% of boys. It is important to remember that this was not a clinical trial but treatment in a busy clinical setting. Attempts to retrieve data from patients not returning as requested were often not successful.

It is possible that the difference between EFH and FH (4.2 +/− 0.5 cm in girls and 5.1 +/− 0.8 cm in boys) would have been smaller with more complete FH data. Results vary considerably in different published reports. These differences may be due to different treatment regimens and inaccuracies in EFH prediction which appears to be more accurate in girls than in boys.4 The mean reduction in FH compared with EFH has been reported as 2.4 cm,5 3.0 cm,6 and 5.2 cm7 in girls and 5.4 cm8 and 9.0 cm9 in boys. Others report far less encouraging results.10

The results reported here suggest a profound change in attitudes to tall stature since around 1990, particularly in girls. Over recent years, compared with previous years, not only fewer girls are being referred, but those being referred are taller. They are less likely to want treatment and the estimated final heights regarded as so unsatisfactory that treatment is requested are taller. Reasons for these changes in attitudes are conjectural. What is regarded as ‘too tall’ has changed over recent decades. The EFH in the last group of girls treated in the current report ranged from 184.4 to 189.3 cm, whereas in Wettenhall's 1975 Australian report,11 treatment was considered if the EFH was >177 cm. Such differences have been developing earlier in other countries.12 In a 1978 American report, the height regarded by girls as too tall was 173 cm in the mid-1960s, rising to 183 cm in the mid 1970s.13

Complications during treatment in this report have been frequent but mild, consistent with other reports,4,7 although 9.4% of treated girls ceased therapy because of excessive weight gain. Girls without FH data had higher BMI SDS before and after treatment than those with FH data. Their failure to return for follow-up could indicate dissatisfaction with a treatment, making a weight problem worse. Formal data on patient satisfaction with treatment and long-term complications are not available in the present study.

However, a major Australian study on 1243 women who had been assessed for tall stature between 1959 and 1993 provides important information on these issues. This examined the relationship between concern about tall stature in adolescence and depression in later life,1 satisfaction in later years with the decision to treat or not to treat14 and long-term effects on fertility.15

Forty-two percent of the treated women expressed dissatisfaction with the decision to treat, whereas 99.1% of untreated women were satisfied with the decision not to treat, regardless of their FH. Reasons for dissatisfaction included not having a say in the decision-making, negative experiences in the assessment and treatment procedures, side-effects during treatment and problems encountered in later life that the participants attributed to their treatment.14

This study also reported that treatment was followed by long-term effects on fertility.15 The treated group had greater difficulty in conceiving than those untreated, although the proportion of women in each group who had ever been pregnant was similar.

Similar long-term concerns apply in boys. In one study, men who had been treated for tall stature as boys had a lower sperm count, lower sperm motility and reduced normal sperm morphology, together with a lower testosterone level than untreated controls.16

The Australian study1 found that the prevalence of major depression in both treated and untreated women was high, compared with the findings of population-based studies, and highlighted the importance of attending to the mental health of adolescents with self-concept and body image concerns.

These findings, although of concern, may not necessarily apply to the patients presented here. Many patients in this study1 received much higher doses of oestrogen than those used in the patients discussed in the current report. In addition, the apparent recent changes in societal attitudes mentioned above could well contribute to a lower prevalence of major depression in tall women as of the present. Data on psychiatric disease in patients reported here are not available.

In summary, a large personal experience in seeing and treating tall children over 25 years is reported. Major changes in attitudes to tallness have led to a substantial reduction in the number of tall girls presenting and treated but not in boys. These changes, and the emerging evidence of interference with both female and male adult fertility by this treatment, are likely to lead to even fewer treatment episodes in the future. The falling numbers of referrals of girls with FTS suggests that the percentage of referrals with an organic cause for tall stature is likely to increase. Therefore, clinicians should be thorough in excluding such causes. Treatment in extreme cases may still be warranted, provided the patient and parents are fully aware of the difficulties of accurate predictions of FH and effectiveness of treatment in an individual, and of recent follow-up data concerning potential long-term side-effects. Psychological evaluation and counselling of the patient may be beneficial. The patient should be involved as much as possible in the decision on whether to treat or not. In general, high-dose sex steroid treatment of tall children should not be undertaken in the future.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

I thank Drs Andrew Cotterill, Mark Harris and Gary Leong for reviewing drafts of the manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References