Endocrine dysfunction and parameters of metabolic syndrome were assessed in 91 patients aged 4.3–32.5 years who underwent allogeneic or autologous BMT in childhood. Final short stature, found in five of the 35 patients who attained final height, was associated with the underlying disease (specifically, Fanconi anemia) (P=0.0013), previous cranial irradiation (P=0.0007), type of conditioning irradiation (P<0.05) and allogeneic BMT (P=0.05). Growth hormone deficiency (n=10) was associated with previous cranial irradiation (P<0.005) and conditioning total body irradiation (P<0.001). Twelve patients had primary hypothyroidism, one had hyperthyroidism and one papillary thyroid carcinoma. Hypothyroidism was associated with neck/mediastinal (P<0.005) and conditioning irradiation (P<0.05). Primary gonadal failure was found in 24 of the mature patients (62.5% females). Hypogonadism was associated with the underlying disease (especially hematological malignancies) (P<0.05), pretransplant treatment (P<0.05), irradiation conditioning (P<0.001), older age (P<0.005) and advanced pubertal stage at BMT (P<0.05). Obesity (body mass index >2 s.d.) was found in 4.4% and type II diabetes and impaired glucose tolerance in 3.3% each. Dyslipidemia was found in 27.9% of the 43 patients tested. These findings emphasize the need for long-term follow-up of endocrine and metabolic parameters in young patients after BMT in order to offer proper treatment and improve quality of life.
The use of BMT, including allogeneic and autologous transplantation of stem cells from bone marrow, peripheral blood and umbilical cord blood, has increased dramatically in recent years. Currently, BMT is applied mainly to treat poor-prognosis leukemia, in addition to specific solid tumors in children, and it is also effective in curing a growing spectrum of hematological, immunological, genetic and metabolic nonhematological diseases.
The number of long-term survivors of BMT is steadily increasing, creating an enlarging pool of children and young adults who are at risk of organ failure owing to the aggressive treatment of their primary disease, the pre-BMT conditioning regimen, or the treatment post stem cell transplantation. Pretransplant conditioning regimens are designed to suppress the immune system in order to promote engraftment, and to eradicate residual malignant cells in patients with cancer. They generally consist of high-dose cyclophosphamide, alone or in combination with total body irradiation (TBI) or high-dose busulfan.
The endocrine organs are well known to be sensitive to both cytotoxic drugs and radiation.1, 2 High rates of skeletal complications,3 growth disturbances,4 and thyroid5, 6 and gonadal dysfunction7 have been described in survivors, although data on the frequency of metabolic syndrome are still sparse.8 The relative risk of these complications is likely to be influenced by the underlying disease, previous treatments, post-BMT treatments and age at BMT.7 In both children and adults, attention is now being focused on these late complications, which may affect health and lower their quality of life.
The aim of the present retrospective long-term analysis was to assess endocrine dysfunction and parameters of metabolic syndrome and their risk factors in survivors of BMT performed during childhood and adolescence attending a referral center in Israel.
Patients and methods
We reviewed the clinical charts of all patients who underwent allogeneic or autologous BMT between 1987 and 2003 and were available for follow-up for at least 1 year after their transplant at the Bone Marrow Transplantation Unit and the Pediatric Institute for Endocrinology and Diabetes of Schneider Children's Medical Center of Israel. Ninety-one disease-free patients, 52 male and 39 female, were eligible for the study. Ages ranged from 4.3 to 32.5 years. Their characteristics are summarized in Table 1a and 1b. Mean age at diagnosis of the underlying disease was 5.6±5.14 years (range, 0.1–18.5). The indications for BMT were as follows: hematological malignancy (44.2%), namely, ALL (n=10), AML (n=22), CML (n=4), and Hodgkin's lymphoma (n=8); solid malignant tumor (28.8%), namely, neuroblastoma (n=16), sarcoma (n=4), meduloblastoma (n=2), Wilms tumor (n=1), germ cell tumor (n=1) and hepatoblastoma (n=1); and other diseases (26.9%), namely, aplastic anemia (n=5), Fanconi anemia (n=4), β-thalassemia major (n=9), Wiskott–Aldrich syndrome (n=2) and hemophagocytic syndrome (n=2). Pretransplant therapy for the underlying disease consisted of chemotherapy alone in 60 patients (65.9%), or combined with irradiation (cranial, cranio-spinal or mediastinal) in 10 (11%); the remaining 21 patients (23.1%) were not treated with chemotherapy or radiation before BMT. Twenty-seven patients received corticosteroid therapy for a period longer than 4 weeks as part of the pretransplant regimen.
The mean age of the whole group at BMT was 7.4±5.2 years (range, 0.6–21.5), and the mean duration of follow-up after BMT was 6.2±3.5 years (range, 1–22.5). Stem cell transplants were of autologous origin in 46 patients (50.5%) and of allogeneic origin, including cord blood, in 45 patients (49.5%).
In 18 patients, the conditioning regimen before BMT consisted of irradiation combined with cyclophosphamide and/or other drugs (busulfan, melphalan, VP-16 or anti-thymocytic globulin). Fourteen patients received TBI at a dose of 12 Gy in six fractions; one received cranial irradiation at a dose of 7 Gy; and three patients with Fanconi anemia received 4–5 Gy thoraco-abdominal irradiation (TAI). The other patients received conditioning chemotherapy (cyclophosphamide and/or busulfan, melphalan, carboplatinum, thiotepa or VP-16) or antithymocytic globulin without irradiation. Most patients received a short course of methotrexate together with cyclosporine for graft-versus-host disease (GVHD) prophylaxis. Patients who received cord blood transplants were given steroids for 4–6 weeks. Chronic GVHD was present in nine patients who underwent allogeneic or cord blood transplantation (20%).
Data on medical and family history, including parental anthropometric measurements and family history of type II diabetes or ischemic heart disease, were obtained for all subjects. Physical examination included measurement of height (with a Harpenden–Holtain stadiometer) and weight, investigation for presence of acanthosis nigricans and determination of pubertal stage (on the basis of breast development in girls and testicular volume in boys) according to the criteria of Marshall and Tanner.9, 10 Physical examination was repeated annually after BMT. Height was expressed as standard deviation score (SDS) for age and gender.11 Short stature was defined as height <−2 s.d. Parental height was measured, and the corrected mid-parental height (target height) was calculated. Final height was defined by bone age ⩾15 years in female patients and 17 years or more in male patients.
Body mass index (BMI) was calculated as the weight in kilograms divided by the square of the height in meters. To compare BMI values across different ages and by sex, the BMI-SDS was calculated. Obesity was defined as BMI >2 s.d.
Failure of spontaneous puberty was defined as the absence of breast development in girls at 13 years of age or more12 and of testicular enlargement in boys at 14 years or more.13 Arrested puberty was defined as lack of advancement of puberty for longer than 1 year, with no advancement to Tanner stage 4–5 after age 16 years. Secondary amenorrhea was defined as the absence of menses for 12 months or longer after menarche.
The annual laboratory assessment included fasting glucose, uric acid, liver enzymes, thyroid function (assessed by thyroid stimulating hormone (TSH) and free thyroxine (FT4)). Patients with elevated levels of liver enzymes (alanine aminotransferase and/or aspartate aminotransferase and gamma glutamyl transferatse ⩾40 U/l) were tested for the presence of hepatitis C and hepatitis B virus.
Forty-three of the 58 patients older than 10 years were tested for fasting triglycerides, total cholesterol, high-density lipoprotein cholesterol (HDL-C) and low-density lipoprotein cholesterol (LDL-C). Dyslipidemia was defined as LDL-C >130 mg/dl and/or HDL-C <35 mg/dl and/or triglycerides >200 mg/dl.
Hypothyroidism was defined as FT4 <10.5 pmol/ml (normal, 10.5–25) with TSH >4.0 mIU/l (normal, 0.5–4.0). Hypogonadism secondary to gonadal dysfunction was defined as an either basal follicle-stimulating hormone (FSH) level of >15 IU/l or basal luteinizing hormone (LH) level of >10 IU/l, with decreased levels of estradiol in girls and testosterone in boys (interpreted according to pubertal stage), or a need for sex hormone replacement. Diabetes mellitus was diagnosed if fasting blood glucose was >126 mg/dl (>6.99 mmol/l) or two random blood glucose measurements were greater than 200 mg/dl (>11.1 mmol/l); impaired glucose tolerance (IGT) was diagnosed if blood glucose levels were 140–199 mg/dl (7.77–11.04 mmol/l), 2 h after an oral glucose tolerance challenge test (OGTT).
Growth hormone (GH) deficiency was defined as a GH level of <10 ng/ml in response to stimulation with clonidine HCl (0.15 mg/m2) or glucagon (30 μg/kg). The tests were performed in patients with short stature (height <−2 s.d.) or reduced growth rate with marked growth deceleration (growth velocity <25th percentile for age).
For the biochemical study, serum glucose was measured by the glucose oxidase colorimetric method using an automated analyzer (Hitachi 917, Roche Diagnostics, Mannheim, Germany). Total cholesterol, triglycerides and HDL-C concentrations were measured by an enzymatic colorimetric method on an automated analyzer (Hitachi 904, Roche Diagnostics). We measured testosterone with a commercial radioimmunoassay (Diagnostic Products Corp., Los Angeles, CA, USA) and TSH, FT4, LH, FSH and GH with a chemiluminescent enzyme immunoassay (Diagnostic Products Corp., Los Angeles, CA, USA).
The statistical analysis was performed using BMDP software.14 Continuous variables were compared across groups using analysis of variance (ANOVA), with Bonferroni correction for multiple comparisons. Associations between discrete variables were analyzed with Pearson's χ2 test or Fisher's exact test, as appropriate. Stepwise logistic regression analysis was used to determine the most significant factors associated with hypogonadism. A P-value of ⩽0.05 was considered significant.
Growth and growth hormone
Mean height-SDS of the whole group at their last evaluation was −0.74±1.23; 16.5% had short stature (height-SDS < −2). Four of the patients with short stature had received corticosteroid therapy for longer than 4 weeks. There was a significant association between short stature and irradiation as part of the conditioning regimen (P<0.05).
Thirty-five patients (38.5%) had attained final adult height at the last evaluation; five (14.3%) of them had short stature. Mean final height was 170.8±9.6 cm in the male patients (n=17) and 161.2±7.2 cm in the female patients (n=18); the mean mid-parental height was 173.4±5.5 and 161.7±6.2 cm, respectively. The difference between the mean final height and the mid-parental height in these patients was not significant.
Final short stature was significantly associated with the underlying disease (specifically Fanconi anemia) (P=0.0013), previous cranial irradiation (P=0.0007), type of conditioning irradiation (especially TBI and TAI) (P=0.023) and allogeneic origin of transplant (P=0.05). It was not associated with use of corticosteroid therapy for the underlying disease, type of conditioning chemotherapy, presence of GVHD or age at BMT (Table 2).
Ten patients (11%) were diagnosed with GH deficiency at a mean age of 11.5±3.2 years (range, 5.5–15) and were treated with recombinant human GH. Mean height-SDS of the patients with GH deficiency was −1.38±1.2, compared to −0.66±1.22 in those without documented GH deficiency. There was no difference between the groups in age at BMT (6.5±4.3 vs 7.52±5.3 years), time since BMT (9.3±6.2 vs 5.8±2.9 years), or origin of transplant. The presence of GH deficiency was significantly associated with corticosteroid therapy for the underlying disease (P<0.005), previous cranial irradiation (P<0.005), and conditioning with TBI (P<0.001). It was not associated with the underlying disease, type of conditioning chemotherapy or presence of GVHD (Table 3). On stepwise logistic regression analysis of the significant variables associated with GH deficiency, conditioning with TBI (odds ratio (OR) 37, 95% confidence interval (CI) 5.94–231) proved to be most predictive. Five of the patients with GH deficiency had attained final adult height at the last evaluation, their mean final height-SDS was −1.13±1.32.
Twelve patients (14%) were diagnosed with primary hypothyroidism (elevated TSH and low FT4 levels) at a mean age of 9.9±6.3 years (range, 0.3–19), after a mean duration of 7 years after BMT. All were treated with l-thyroxine. None had elevated levels of antiperoxidase or antithyroglobulin antibodies. There were no cases of compensated hypothyroidism (elevated TSH level with a normal FT4 level). One patient was diagnosed with hyperthyroidism 1.5 years after BMT and is being treated with propylthiouracil. This patient did not have GVHD.
There was no difference in the prevalence of hypothyroidism between male and female patients, and no difference in age at BMT and follow-up interval after BMT between patients with and without hypothyroidism (mean age at BMT: 7.9±5.3 vs 7±5.1 years, respectively; mean interval after BMT: 7±4.1 vs 6.1±3.5 years, respectively). Hypothyroidism was significantly associated with neck/mediastinal irradiation therapy for the underlying disease (P<0.005) and irradiation as part of the conditioning regimen (P<0.05), but not with the underlying disease, corticosteroid therapy for the underlying disease, type of conditioning chemotherapy, origin of transplant or presence of GVHD.
One patient was diagnosed with papillary thyroid carcinoma at age 30 years (15.5 years after BMT with TBI conditioning). Treatment consisted of total thyroidectomy and radio-iodine ablation therapy.
Gonadal function and puberty
At the time of BMT, 29 of the 39 girls were prepubertal, three had begun puberty and seven were Tanner stage 5 (Table 1a). By the time of the last evaluation, 28 girls had either reached the age of 13 years or begun spontaneous puberty at a mean age of 10.6±1.1 years. Fifteen of them (53.6%) were diagnosed with hypogonadism at a mean age of 14.7±2.7 years, including three who failed to attain puberty, three with arrested puberty and nine with secondary amenorrhea and evidence of gonadal dysfunction. Primary gonadal failure was accompanied in all patients by abnormally elevated levels of FSH and/or LH. All these patients received hormone replacement therapy.
At the time of BMT, 40 of the 52 boys were prepubertal, six had begun puberty and six had attained Tanner stage 5 (Table 1a). By the time of the last follow-up visit, 31 boys had either reached the age of 14 years or begun spontaneous puberty at a mean age of 11.6±1.4 years. Nine (29%) of the 31 patients were found to have biochemical or clinical evidence of hypogonadism at a mean age of 15.2±2.2 years, including two who failed to attain puberty, five with arrested puberty and two with evidence of gonadal dysfunction. Sperm evaluations were not routinely performed. Male patients with inappropriately low testosterone levels were treated with monthly testosterone enanthate injections.
Female patients accounted for 62.5% of the patients with primary gonadal failure. In both male and female patients, there was a significant association between presence of hypogonadism and the underlying disease, especially hematological malignancies in female patients (P<0.05), and its treatment (P<0.05), receipt of irradiation therapy as part of the conditioning regimen (P<0.001) and advanced pubertal stage at BMT (P<0.05). Patients who had hypogonadism were significantly older at BMT than those who did not (10.9±4.8 vs 7.2±4.6 years, P<0.005). We did not find a significant association between hypogonadism and origin of transplant (autologous vs allogeneic) or type of conditioning chemotherapy (Table 4), even when we compared patients receiving busulfan-containing and non-busulfan regimens. On stepwise logistic regression analysis of the significant variables associated with hypogonadism, advanced Tanner stage at BMT (OR 1.56, 95% CI 1.03–2.37) and conditioning irradiation (OR 16.4, 95% CI 2.92–92.1) proved to be most predictive.
Diabetes mellitus and parameters of metabolic syndrome
Mean BMI-SDS of the whole group at their last evaluation was 0.26±1.06; four patients (4.4%) had a BMI-SDS of >2. Mean BMI of the mothers was 26.7±5.8 and of the fathers 26.5±3.0. There was no difference in age at BMT or duration as BMT between patients with a BMI-SDS of more or less than 2. In addition, no association was found between BMI-SDS at the last evaluation and the underlying disease or its treatment, use of corticosteroid therapy for more than 4 weeks as a pretransplant treatment, or type of BMT conditioning. Only two patients (2.2%) had acanthosis nigricans.
Dyslipidemia was found in 27.9% of the 43 patients tested: LDL-C level was >130 mg/dl in nine patients (20.9%); HDL-C level was <35 mg/dl in four patients (9.3%) and triglyceride level was >200 mg/dl in five patients (11.6%). Hyperuricemia (uric acid >7 mg/dl) was found in six patients (6.6%).
Three patients (3.3%) were diagnosed with type II diabetes mellitus and three (3.3%) with IGT at a mean age of 21.5±6.1 years (range, 15–28). A BMI-SDS of <2 was noted in all three patients with diabetes and in two of the three patients with IGT. Age at BMT was similar in patients with and without diabetes, although in the diabetic group, the mean duration since BMT was longer (18.7±3.4 vs 5.8±2.7 years, P<0.05). Presence of diabetes or IGT was not associated with origin of the transplant. Two patients with diabetes and one with IGT also had elevated liver enzymes and were positive for hepatitis C infection. All the patients with diabetes were irradiated as part of the conditioning therapy and received corticosteroids before transplantation for more than 4 weeks. Neither diabetes nor IGT was associated with the underlying disease or its treatment, presence of chronic GVHD, parental BMI or family history of type II diabetes.
Treatment protocols for BMT have varied throughout the years, and new indications for its use are increasing. Because BMT is generally performed in young patients with a long life expectancy, its long-term endocrine complications must be assessed. To our knowledge, this is the first review of endocrine complications in young patients who underwent BMT in Israel.
The 14.3% rate of final short stature in our patients is lower than the 30% (out of 45 patients) reported by Legault and Bonny5 and the 25% (out of 48 patients) reported by Bakker et al.,15 but it is higher than the 4% (out of 28 patients) found by Cohen et al.16 This variability in final height outcome may be attributable to the different patient ages at BMT, as the risk of impaired growth is highest in the youngest children.17 However, we failed to note an association of younger age at BMT with height loss. In our group, the mean final height of the patients was not significantly different from their mid-parental height.
It is likely that a multitude of factors interact to cause impaired linear growth following BMT in children and adolescents. Potential mechanisms include the direct impact of combination chemotherapy and/or radiation on the skeleton with skeletal dysplasia, corticosteroid therapy, GVHD, radiation-induced GH deficiency and hypothyroidism. The situation is further complicated in patients receiving cranial irradiation for the underlying disease.
We found that disturbed growth was significantly associated with conditioning irradiation, as previously reported,18, 19 but not with the type of non-TBI conditioning chemotherapy, similar to findings in most of the earlier studies.20, 21 However, Bakker et al.15 noted an association of impaired growth with busulfan/cyclophosphamide conditioning. They speculated that the ‘toxic’ actions of corticosteroids on the epiphyseal growth plate can persist in part even after their administration is ceased. Nevertheless, in our cohort, final short stature was not associated with use of corticosteroid therapy, perhaps owing to the relatively short term of the therapy.
We did not find an association of short stature with GVHD, as opposed to Sanders et al.19 who explained their positive finding by both the disease process itself and the prolonged steroid therapy it warrants. We also failed to note an association of younger age at BMT with height loss, in contradiction to previous studies.16, 17 This discrepancy may be attributed to the low percentage of our patients (38.5%) who had attained final height by the time of the study.
In our cohort, final short stature was associated with the underlying disease, especially Fanconi anemia, which is known to have a high incidence of short stature. Thus, is possible that growth after BMT in patients with Fanconi anemia was affected by their constitutional factors. We also found a significant association between final short stature and allogeneic origin of the transplant, as previously described.5, 15 However, this might be explained by the fact that all our patients with Fanconi anemia underwent allogeneic BMT.
Corticosteroid excess may blunt GH secretion.22 We found a significant association between GH deficiency and corticosteroid therapy, administered as part of the treatment protocol of the underlying disease, although the GH deficiency was diagnosed at least 3 years after corticosteroids were discontinued. Many studies have reported a significant incidence of GH deficiency following cranial irradiation,19, 23 and also after TBI.24 This association was documented in our group as well. Growth hormone treatment after BMT has usually been reported to result in increase in growth velocity, albeit a smaller one than in children with idiopathic GH deficiency, which is probably caused by the lesions induced by TBI.25 All our patients diagnosed with GH deficiency were treated with recombinant growth hormone (rGH), which had a beneficial effect on growth. Based on our results, we suggest that rGH be administered to children with a decreased growth rate and low GH values on stimulation tests, presuming that other hormones necessary for optimal growth are adequately replaced.
Thyroid dysfunction has been reported to occur in both children26, 27 and adults after TBI, with a greater risk among younger patients.28 The most common type of thyroid dysfunction in long-term survivors of BMT is subclinical compensated or overt hypothyroidism, whose frequency seems to be related to the use of the conditioning therapy, including TBI. Although hypothyroidism is usually a relatively early complication of BMT, it can also manifest years later.
In our group, 14% of the patients had primary hypothyroidism, similar to the incidence reported by Bouland et al.29 and Legault and Bonny.5 Mean age at diagnosis was 9.9±6.3 years. We found a significant association of hypothyroidism with neck/mediastinal irradiation for the underlying disease and irradiation as part of the conditioning, but not with age at BMT, time since BMT, or type of conditioning chemotherapy. Au et al.30 reported the occurrence of autoimmune thyroid dysfunction after hematopoietic stem cell transplantation, but none of our patients with hypothyroidism had elevated levels of antiperoxidase or antithyroglobulin antibodies. This discrepancy may be explained by the different human lymphocyte antigens associated with autoimmune thyroid diseases in different populations. Annual testing is recommended, as many of these patients are relatively asymptomatic.
The risk of thyroid tumors after irradiation is known to be dose related.31 Furthermore, the thyroid gland is especially sensitive to the effects of irradiation at a very young age.32 The latency period between thyroid irradiation and the clinical presentation of a thyroid tumor may be many years. Socie et al.32 reported an incidence of 0.2% of thyroid carcinoma among 3182 children after BMT, with a strong relationship between the occurrence of thyroid cancer and age at BMT. Ishiguro et al.28 reported a high incidence of thyroid adenoma after BMT. However, we found only one case of thyroid tumor (papillary carcinoma) in a patient who underwent BMT at age 14.5 years. This difference may have been due to our previous policy of performing thyroid ultrasound only in patients with palpable thyroid nodules or suspicious consistency of the thyroid gland, and in patients treated with neck or mediastinal irradiation. By contrast, Ishiguro et al.28 performed thyroid ultrasound annually in many of their patients. Our new suggested surveillance approach includes thyroid palpation with performance of thyroid ultrasound in patients with a suspicious consistency of the thyroid gland, or annually in patients after neck, thoraco-mediastinal irradiation or TBI.
Puberty and gonadal function
Pubertal disturbances after BMT are caused by damage to the ‘central unit’ of the hypothalamus–pituitary region or ‘peripheral’ damage to the gonads. The central unit can be functionally damaged by cranial irradiation, leading to impaired gonadotropin secretion. Gonadal damage is also a well-known side effect of alkylating agents, such as busulfan, cyclophosphamide and melphalan.20 Primary gonadal failure was the presenting sign in all of our patients with hypogonadism, and a high percentage of these patients were female (62.5 vs 37.5%). The risk of gonadal failure is known to increase with cumulative doses of gonadotoxic therapies (radiation and/or alkylating agents). However, the relative contribution of high-dose conditioning chemotherapy to gonadal failure in our patients was difficult to establish, because the majority who were treated for malignant disease received chemotherapy before BMT.
In both the male and female patients, the incidence of hypogonadism was higher in those with hematological malignancies than in those with other underlying diseases, perhaps owing to the specific treatment required.
Young boys and male adolescents who receive cyclophosphamide alone or TBI-based preparative regimens appear to retain normal Leydig cell function with normal plasma levels of testosterone. However, they have evidence of germ cell dysfunction, with increased plasma levels of FSH and a reduced testicular volume, which correlates with impaired spermatogenesis.33 The chemotherapy-induced germ cell damage may be more common in those treated during or after puberty than in prepubertal males.
Pubertal girls treated with busulfan and cyclophosphamide are at a very high risk of developing ovarian failure.5, 15, 34 The outcome of ovarian function following TBI appears to be determined by the age of the patient at the time of irradiation. Previous data indicate that approximately 50% of prepubertal girls given TBI will enter puberty spontaneously and achieve menarche at a normal age,35 whereas almost all female patients who are more than 12 years old at the time of TBI have ovarian failure, probably because of the decreased number of primordial follicles.35 We found that hypogonadism was significantly associated with conditioning with TBI or TAI, but not with the type of conditioning chemotherapy, perhaps owing to the high percentage of patients who received alkylating agents. Older age and advanced pubertal stage at BMT were both significantly associated with hypogonadism, although Teinturier et al.36 found that high-dose busulfan is a major cause of ovarian failure, even when given in the prepubertal period without TBI or abdominal or pelvic radiotherapy.
Diabetes and parameters of the metabolic syndrome
Disturbances of insulin-related metabolism occur less often than other endocrine disturbances after BMT. In our series, the rates of type II diabetes and IGT were 3.3% each, and of dyslipidemia 27.9%. In only 4.4% of our patients BMI-SDS was >2 s.d. Accordingly, Lorini et al.37 reported that all of their 34 patients undergoing BMT for hematological malignancies, of whom 24 had received a combination of cytotoxic drugs and TBI, had normal glucose levels on intravenous glucose tolerance test and a normal HbA1C concentration, with significantly elevated levels of insulin. The highest insulin levels were noted in the patients who received preparation with both cytotoxic drugs and TBI.
Taskinen et al.,8 in a series of 23 long-term survivors of BMT (median age 20 years), reported a higher prevalence of insulin resistance (52%), IGT (26%), type II diabetes (17%) and signs of the metabolic syndrome (39%) even in those of normal weight and young age.
In our group, there was a significant association between presence of diabetes and irradiation as part of the conditioning therapy, but the mechanism is unclear. This association was not demonstrated in the series of Taskinen et al.8 The latter authors also found that the frequency of insulin resistance increased with the time since BMT and presence of chronic GVHD, whereas we failed to note a significant association between the development of diabetes or IGT and age at BMT, time since BMT, presence of chronic GVHD or family history of type II diabetes.
It has long been recognized that steroids can induce diabetes usually during the treatment period.38, 39 However, in our series, corticosteroid therapy, in all the patients with diabetes and in one with IGT, was administered before transplantation, several years before development of the disease. Two of our patients with diabetes and one with IGT had hepatitis C infection, and this may have been the cause of the impaired glucose metabolism as previously reported,40, 41 and not the BMT.
This study has some limitations. First, the findings are limited somewhat by the relatively small number of patients who achieved final adult height at the latest evaluation, and by the many patients who were still prepubertal at the last assessment. The relatively young age of our patients may also have had a significant impact on the incidence of metabolic disorders. It is likely that a larger cohort with longer follow-up would provide more useful information. The second limitation is the study's retrospective design, as the clinical and laboratory assessments were not optimal. Thus, the differences in the rates of diabetes and IGT from previous reports may be attributable to our performance of OGTT only in patients with impaired fasting glucose, randomly elevated levels of glucose, obesity, acanthosis nigricans or clinical signs such as polyuria or polydipsia. Third, we did not perform a routine annual thyroid ultrasound scan in all patients to detect structural changes or nonpalpable nodules. Finally, GH deficiency was diagnosed by stimulation tests in 11% of our patients. However, the tests were performed only in patients exhibiting short stature or reduced growth rate with marked growth deceleration. Thus, the real prevalence of these disturbances may be underestimated.
In summary, our study suggests that patients after BMT tend to have growth disorders, thyroid and gonadal dysfunctions, and some laboratory features of the metabolic syndrome. Even though final cure is the ultimate goal of BMT, treatment regimens will probably have long-term endocrine implications. Therefore, patients require life-long follow-up to detect and treat any endocrine and metabolic dysfunctions and to maintain their quality of life.
Shalet SM . Endocrine consequences of treatment of malignant disease. Arch Dis Child 1989; 64: 1635–1641.
Kolb HJ, Bender-Gotze C . Late complications after allogeneic bone marrow transplantation for leukemia. Bone Marrow Transplant 1990; 6: 61–72.
Kashyap A, Kadeel F, Yamauchi D, Palmer JM, Niland JC, Molina A et al. Effects of allogeneic bone marrow transplantation on recipient bone mineral density: a prospective study. Biol Blood Marrow Transplant 2000; 6: 344–351.
Brauner R, Fontoura M, Zucker JM, Devergie A, Souberbielle JC, Prevot-Saucet C et al. Growth and growth hormone secretion after bone marrow transplantation. Arch Dis Child 1993; 68: 458–463.
Legault L, Bonny Y . Endocrine complications of bone marrow transplantation in children. Pediatr Transplant 1999; 3: 60–66.
Toubert ME, Socie G, Gluckman E, Aractingi S, Esperou H, Devergie A et al. Short- and long term follow-up of thyroid dysfunction after allogeneic bone marrow transplantation without the use of preparative total body irradiation. Br J Haematol 1997; 98: 453–457.
Shalet SM, Didi M, Oglivy-Stuart AL, Schulga J, Donaldson MD . Growth and endocrine function after bone marrow transplantation (Review). Clin Endocrinol 1995; 42: 333–339.
Taskinen M, Saarinen-Pinkala UM, Hovi L, Lipsanen-Nyman M . Impaired glucose tolerance and dyslipidemia as late effects after bone-marrow transplantation in childhood. Lancet 2000; 16: 993–997.
Marshall WA, Tanner JM . Variation in the pattern of pubertal changes in girls. Arch Dis Child 1969; 49: 291–303.
Marshall WA, Tanner JM . Variations in the pattern of pubertal changes in boys. Arch Dis Child 1970; 45: 13–23.
Kuczmarski RJ, Ogden CL, Grummer-Strawn LM . CDC Growth Charts: United States Advance Data From Vital and Health Statistics, No. 314. National Center for Health Statistics: Hyattsville, MD, 2000.
Rosenfield R . Puberty in the female and its disorders. In: Sperling MA (ed). Pediatric Endocrinology, 2nd edn. Elsevier Science: Pennsylvania, 2002, p 494.
Styne DM . The testes: disorders of sexual differentiation and puberty in the male. In: Sperling MA (ed). Pediatric Endocrinology, 2nd edn. Elsevier Science: Pennsylvania, 2002, p 599.
Dixon WJ . BMDP Statistical Software. University of California Press: Los Angeles, 1993.
Bakker B, Oostdijk W, Bresters D, Walenkamp MJ, Vossen JM, Wit JM . Disturbances of growth and endocrine function after busulphan-based conditioning for haematopoietic stem cell transplantation during infancy and childhood. Bone Marrow Transplant 2004; 33: 1049–1056.
Cohen A, Rovelli A, Van-Lint MT, Uderzo C, Morchio A, Pezzini C et al. Final height of patients who underwent bone marrow transplantation during childhood. Arch Dis Child 1996; 74: 437–440.
Frisk P, Arvidson J, Gustafsson J, Lonnerholm G . Pubertal development and final height after autologous bone marrow transplantation for acute lymphoblastic leukemia. Bone Marrow Transplant 2004; 33: 205–209.
Giorgiani G, Bozzola M, Locatelli F, Picco P, Zecca M, Cisternino M et al. Role of busulfan and total body irradiation on growth of prepubertal children receiving bone marrow transplantation and results of treatment with recombinant human growth hormone. Blood 1995; 86: 825–831.
Sanders JE, Pritchard S, Mahoney P, Amos D, Buckner CD, Witherspoon RP et al. Growth and development following marrow transplantation for leukemia. Blood 1986; 68: 1129–1135.
Shankar SM, Bunin NJ, Moshang T . Growth in children undergoing bone marrow transplantation after busulphan and cylophosphamide conditioning. J Pediatr Hematol Oncol 1996; 18: 362–366.
Afify Z, Shaw PJ, Clavano-Harding A, Cowell CT . Growth and endocrine function in children with acute myeloid leukemia after bone marrow transplantation using busulphan/cylophosphamide. Bone Marrow Transplant 2000; 25: 1087–1092.
Thompson RG, Rodriguez A, Kowarski A, Blizzard RM . Growth hormone: metabolic clearance rates, integrated concentrations, and production rates in normal adults and effect of prednisone. J Clin Invest 1972; 51: 3193–3199.
Brauner R, Rappaport R, Prevot C, Czernichow P, Zucker JM, Bataini P et al. A prospective study of the development of growth hormone deficiency in children given cranial irradiation, and its relation to statural growth. J Clin Endocrinol Metab 1989; 68: 346–351.
Ogilvy-Stuart AL, Clark DJ, Wallace WH, Wallace WH, Gibson BE, Stevens RF et al. Endocrine deficit after fractionated total body irradiation. Arch Dis Child 1992; 67: 1107–1110.
Brauner R, Adan L, Souberbielle JC, Esperou H, Michon J, Devergie A et al. Contribution of growth hormone deficiency to the growth failure that follows bone marrow transplantation. J Pediatr 1997; 130: 785–792.
Katsanis E, Shapiro RS, Robison LL, Haake RJ, Kim T, Pescovitz OH et al. Thyroid dysfunction following bone marrow transplantation: long-term follow-up of 80 pediatric patients. Bone Marrow Transplant 1990; 5: 335–340.
Borgstrom B, Bolme P . Thyroid function in children after allogeneic bone marrow transplantation. Bone Marrow Transplant 1994; 13: 59–64.
Ishiguro H, Yasuda Y, Tomita Y, Shinagawa T, Shimizu T, Morimoto T et al. Long-term follow-up of thyroid function in patients who received bone marrow transplantation during childhood and adolescence. J Clin Endocrinol Metab 2004; 89: 5981–5986.
Bouland F, Bromley M, Black P, Heller G, Sarafoglou K, Gillio A et al. Thyroid dysfunction following bone marrow transplantation using hyperfractionated radiation. Bone Marrow Transplant 1995; 15: 71–76.
Au WY, Lie AK, Kung AW, Liang R, Hawkins BR, Kwong YL . Autoimmune thyroid dysfunction after hematopoietic stem cell transplantation. Bone Marrow Transplant 2005; 35: 383–388.
Sklar C, Whitton J, Mertens A, Stovall M, Green D, Marina N et al. Abnormalities of the thyroid in survivors of Hodgkin's disease: data from the Childhood Cancer Survivors Study. J Clin Endocrinol Metab 2000; 85: 3227–3232.
Socie G, Curtis RE, Deeg HJ, Sobocinski KA, Filipovich AH, Travis LB et al. New malignant diseases after allogeneic bone marrow transplantation for childhood acute leukemia. J Clin Oncol 2000; 18: 348–357.
Sklar CA, Kim TH, Ramsay NK . Testicular function following bone marrow transplantation performed during or after puberty. Cancer 1984; 53: 1498–1501.
Sarafoglou K, Boulad F, Gillio A, Sklar C . Gonadal function after bone marrow transplantation for acute leukemia during childhood. J Pediatr 1997; 130: 210–216.
Sanders JE . The impact of marrow transplant preparative regimens on subsequent growth and development. Semin Hematol 1991; 28: 244–249.
Teinturier C, Hartmann O, Valteau-Couanet D, Benhamou E, Bougneres PF . Ovarian function after autologous bone marrow transplantation in childhood: high-dose busulfan is a major cause of ovarian failure. Bone Marrow Transplant 1998; 22: 989–994.
Lorini R, Cortona L, Scaramuzza A, De Stefano P, Locatelli F, Bonetti F et al. Hyperinsulinemia in children and adolescents after bone marrow transplantation. Bone Marrow Transplant 1995; 15: 873–877.
Yang JY, Cui XL, He XJ . Non-ketotic hyperosmolar coma complicating steroid treatment in childhood nephrosis. Pediatr Nephrol 1995; 9: 621–626.
Raul Ariza AC, Barile-Fabris LA, Frati-Munari AC, Baltazar- Montufan P . Risk factors for steroid diabetes in rheumatic patients. Arch Med Res 1998; 29: 259–262.
Zein CO, Levy C, Basu A, Zein NN . Chronic hepatitis C and type II diabetes mellitus: a prospective cross-sectional study. Am J Gastroenterol 2005; 100: 48–55.
Kawaguchi T, Nagao Y, Tanaka K, Ide T, Harada M, Kumashiro R et al. Causal relationship between hepatitis C virus core and the development of type 2 diabetes mellitus in a hepatitis C virus hyperendemic area: a pilot study. Int J Mol Med 2005; 16: 109–114.
We thank Pearl Lilos for the statistical analysis and Gloria Ginzach for the editorial assistance.
About this article
Cite this article
Shalitin, S., Phillip, M., Stein, J. et al. Endocrine dysfunction and parameters of the metabolic syndrome after bone marrow transplantation during childhood and adolescence. Bone Marrow Transplant 37, 1109–1117 (2006). https://doi.org/10.1038/sj.bmt.1705374
- thyroid gland
A systematic approach to the endocrine care of survivors of pediatric hematopoietic stem cell transplantation
Cancer and Metastasis Reviews (2020)
Pubertal outcomes of children transplanted with allogeneic stem cells after myeloablative total body irradiation or busulfan: Influence of age and sex is confirmed, while a role of chronic graft‐versus‐host disease in delayed puberty onset is revealed
Pediatric Transplantation (2020)
The challenge of long‐term follow‐up of survivors of childhood acute leukemia after hematopoietic stem cell transplantation in resource‐limited countries: A single‐center report from Brazil
Pediatric Transplantation (2020)
Delineating the role of FANCA in glucose-stimulated insulin secretion in β cells through its protein interactome
PLOS ONE (2019)
Endocrine Practice (2019)