RESULTS

Evaluations of Growth Retardation

Body weight was 21 kg (-3.31 SDS) and height 118 cm (-2.35 SDS). The status of sexual maturation was prepubertal. Bone age was 3 years younger than the chronological age. GH provocative tests with insulin-induced hypoglycemia, clonidine, and levodopa showed peak GH levels of 2.1, 3.2, and 1.8 ng/mL, respectively ( Table 2 ). All of these findings were compatible with GH deficiency.

TABLE 2 -- Table 2. Changes of Clinical and Laboratory Findings Pertinent to Growth Retardation by Recombinant Human Growth Hormone Therapy

	Before Treatment	rhGH + K+	K+ Only
Height	118 cm (-2.35 SDS)	124 cm (-2.15)	125.8 cm (-2.39)
Weight	21 kg (-3.31 SDS)	25 kg (-0.17)	26 kg (-1.55)
Chronological age	9 yr 10 mo	10 yr 4 mo	10 yr 10 mo
Bone age	7 ± 1 yr	7 ± 1 yr	8 ± 1 yr
Sexual maturity rate (Tanner)
Pubic hair	Stage 1	Stage 1	Stage 1
Penis	Stage 1	Stage 1	Stage 1
Testicular volume	2 cc (Rt, Lt)	2 cc (Rt, Lt)	2 cc (Rt, Lt)
Peak serum GH level provoked with
Insulin-induced hypoglycemia	2.1 ng/mL	--	--
Clonidine	3.2 ng/mL	--	--
Levodopa	1.8 ng/mL	--	--
Serum IGF-I	55 ng/mL	129 ng/mL	68 ng/mL
NOTE. During the first 6 months of therapy, combined recombinant human growth hormone and potassium supplement (rhGH + K+ ) were given. During the next 6 months of therapy, only a potassium supplement (K+ ) was given.
Abbreviations: GH, growth hormone; IGF-I, insulin-like growth factor-I.

Efficacy of rhGH Therapy

During the first 6 months, potassium was given as potassium chloride, 3 mEq/kg/d. And therapy with rhGH, 0.1 IU/kg/d, was combined with the above. Thereafter, therapy with rhGH was stopped, and only potassium was given during the next 6 months.

Our patient exhibited an excellent effect on short-term linear growth with rhGH therapy together with potassium supplement. Yearly height increment increased threefold during rhGH therapy compared with that of pretreatment. This increment is almost twofold that of normal children at his age. In a review of literature, it was noted that other forms of treatments kept growth velocity within normal range or slightly higher than normal range.^[4] ^{[9] [10] [15] [16]} Therefore, the effect of rhGH therapy on growth velocity in this child exceeded that of other treatments.

Changes in Blood Electrolytes and Urinary Excretion Rates After rhGH Therapy and After Its Withdrawal

Serum magnesium level normalized after rhGH therapy (1.3 to 2.4 mg/dL) and decreased to 1.7 mg/dL after withdrawal of rhGH. Urinary magnesium excretion rates (urinary magnesium-to-creatinine ratio) were not changed by rhGH therapy ( Table 1 ).

Serum phosphorus level rose sharply after rhGH therapy when compared with the level before rhGH therapy (2.5 to 4.3 mg/dL). After withdrawal of rhGH, it decreased to 3.0 mg/dL. Changes of serum calcium levels by rhGH therapy were not marked. Urinary calcium excretion rates (urinary calcium-to-creatinine ratio) also were not changed markedly by rhGH therapy ( Table 1 ).

Immunoactive parathyroid hormone (PTH) levels before, after, and during withdrawal of rhGH were 25, 13, and 38 pg/mL, respectively. However, they were all within normal ranges ( Table 1 ).

DISCUSSION

In many children with Bartter's syndrome and its variants, body growth was reported to be retarded,^{[4] [5] [6]} but only a few studies analyzed growth retardation in this disease. In 1979, Simopoulos^[4] concluded that conventional treatments such as potassium or magnesium supplementation do not positively influence linear growth, bone age, or weight gain in children with Bartter's syndrome. Other reports suggest that some beneficial effect on short-term growth may be obtained by potassium supplementation, especially in mild forms of the disease. ^[17] Good growth velocity has been reported with spironolactone.^{[4] [9]} More recently, the cyclooxygenase inhibitor, indomethacin, has appeared to improve the growth of older children with Bartter's syndrome.^[15]

Indomethacin therapy seems to yield quite good results in terms of growth. Seidel et al^[16] hypothesized that indomethacin supports skeletal growth by suppression of osteolysis and by reduction of calciuria. However, other reports suggest that bone resorption leading to hypercalciuria is indomethacin resistant.^[18] Proesmans et al^[10] reported a patient with Bartter's syndrome in whom growth normalized with indomethacin therapy, although the hypercalciuria was not corrected, and nephrocalcinosis was observed by ultrasonography at the age of 19 years.^[10] Therefore, growth retardation is not directly related to bone resorption, but it seems to be caused by a pathogenesis not known as yet. Compared with indomethacin, rhGH therapy seems to be a better treatment in terms of growth. rhGH therapy has been known to be safe and to have a very excellent effect on long-term growth in children with short stature with various causes since the 1980s.

The pathogenesis of growth retardation in Bartter's syndrome and Gitelman's syndrome has not been clearly defined. Simopoulous concluded that potassium per se is not related to growth retardation in Bartter's syndrome.^[4] The disturbance of calcium metabolism related to osteopenia may be another cause of growth retardation.^[18] Some degree of malnutrition may contribute to growth retardation in many patients with Bartter's syndrome.^[16] Recently, GH deficiency has been reported in patients with Bartter-like diseases. ^{[7] [8]} We now report that growth hormone was deficient in a short child with Gitelman's syndrome. Therefore, we recommend that GH provocative tests should be done in short children with Bartter's syndrome or Gitelman's syndrome showing delayed bone age to evaluate their GH status. We suggest that it is also necessary for future study to gather a cohort of these patients and to measure GH levels.

In summary, we evaluated the status of GH secretion in a short child with Gitelman's syndrome and found that GH was deficient and rhGH therapy markedly improved this patient's growth rate and also restored the serum magnesium level to normal. We believe that rhGH therapy should be considered in short children with Gitelman's syndrome or Bartter's syndrome that are resistant to conventional therapies in terms of growth or that reveal limited effect on growth velocity with conventional therapies such as potassium supplement, spironolactone, or indomethacin.

ACKNOWLEDGMENT

We thank Dr GA Quamme, University of British Columbia, Vancouver, Canada, for his valuable advice.

REFERENCES