Long-term reduction in bone mass after severe burn injury in children

Long-term reduction in b o n e mass after severe burn injury in children G o r d o n L. Klein, MD, David N. Herndon, MD, Craig B. L a n g m a n , MD, Thomas C. Rutan, MSN, William E. Y o u n g , MS, G r e g o r y P e m b l e t o n , CNMT, Martin Nusynowitz, MD, J o s e p h L. Barnett, BSN, Lyle D. Broemeling, PhD, D a w n E. Sailer, MS, a n d Robert L. M c C a u l e y , MD From the Departments of Pediatrics, Surgery, and Nuclear Medicine, University of Texas Medical Branch and ShrinersBurns Institute, Galveston, and the Division of Nephrology Mineral Metabolism Laboratory, Children's Memorial Hospital and Northwestern University Medical School, Chicago, illinois

Objective: Because burn victims are at risk of having bone loss, a cross-sectional study was undertaken to determine whether severe burn injury had acute and longterm effects on bone mass or on the incidence of fractures in children. Methods: Dual-energy x-ray absorptiometry of the lumbar portion of the spine was performed on 68 children: 16 moderately burned (15% to 36% of total body surface area) and 52 age-matched severely burned (>40% of total body surface area). Twenty-two severely burned children were hospitalized and studied within 8 weeks of their burn, and 30 others were studied approximately 5 years after discharge. In the severely burned group, both hospitalized and discharged, serum and urine were analyzed for calcium, phosphorus, intact parathyroid hormone, osteocalcin, and type I collagen telOpeptide. Results: Sixty percent of severely burned patients had age-related z scores for bone density less than - I , and 27% of severely burned patients had age-related zscores for bone density less than - 2 (p <0.005, for each). In the moderately burned group, 31% of patients had z scores less than - I (p <0.005 vs normal distribution), but only 6% had z scores less than - 2 (p value not significant). There was evidence of increased incidence of fractures after discharge in the severely burned patients. Biochemical studies were compatible with a reduction in bone formation and an increase in resorption initially, and with a long-term persistence of low formation. Conclusion: We conclude that acute burn injury leads to profound and long-term bone loss, which may adversely affect peak bone mass accumulation. (J PEDIATR 1995;126:252-6)

Supported in part by a grant from the Shriners Hospitals for Crippled Children (SHCC 15877) (Dr. Klein) and an unrestricted grant from the Otto Sprague Memorial Fund, and in part by the General Clinical Research Center (grant RR00048), National Institutes of Health (Dr. Langman). Presented in part at the Third Sino-Americanconference on Burns and Trauma, Canton, China, Aug. 16 to 19, 1993, and at the Fifteenth Annual Meeting of the American Society for Bone and Mineral Research, Tampa, Fla., Sept. 18 to 22, 1993. Submitted for publication April 8, 1994; accepted Aug. 18,1994. Reprint requests: Gordon L. Klein, MD, Pediatric Gastr*oenterology Division,Children's Hospital, Rm C353, University of Texas Medical Branch, Galveston, TX 77555-0352. Copyright ® 1995 by Mosby-Year Book, Inc. 0022-3476/95/$3.00 + 0 9/20/62556 252

The long-term effects of major thermal injuries on growth and development of children are now receiving significant attention. Rutan and Herndon 1 reported significant growth delays in children for up to 3 years after survival from thermal injury, and permanent stunting of linear growth. However, little attention has been directed to the effects of such injury on bone metabolism. Recently we identified the acute onset of alterations in bone turnover associated with burn injury in young adults. As demonstrated by bone biopsy, this abnormality consisted of a marked reduction in bone formation combined with a decreased area of unmineralizedosteoid; changes in bone resorption were inconsistent.2 The disproportionate reduction in bone formation

The Journal o f Pediatrics Volume 126, Number 2

Klein et al.

suggests that burned patients are at risk of having bone mineral loss within the first 2 months after burn injury. The pathogenesis of the lesion may include such factors as immobilization,2 increased endogenous production of corticosteroids,2 aluminum loading,2 and the production of bone resorptivc cytokines such as interleukin- 13 and interleukin-6.4, 5 TBSA

Total body surface area

[

Despite the early presence of this burn-associated bone abnormality, neither its duration nor its clinical significance is known. Therefore our present study was designed to determine whether a deficit in bone mass persisted after recuperation from burn injury, and whether the reduction in bone mass was associated with an increase in fractures in this population. METHODS Patients. We studied 68 children and adolescents recruited from the inpatient acute care unit or the reconstructive surgery clinic at the Shriners Burns Institute, Galveston, Tex., from November 1992 through April 1994. The patients were divided initially into two groups on the basis of the severity of thermal injury.6, 7 Burns in patients with less than 40% of their total body surface area burned were classified as moderate (range, 15% to 36%; n = 16), and burns in those with burns >40% of TBSA were classified as severe (range, 40% to 100%; n = 52). Clinical features of all patients are shown in Table I. The Shriners Burns Institute, Galveston Unit, is a referral hospital whose patients come from the southwestern United States. The ethnic backgrounds of the study patients were 60% white, 30% Hispanic, and 10% black. Severely burned patients were subdivided by inpatient or outpatient status. The inpatients (n = 22) were studied during their hospitalization, which usually was during the first 8 weeks after the burn; the outpatients (n = 30) were studied after discharge when they returned to the reconstructive surgery clinic (Table I). We matched sequentially, during a 17month period, the ages and percentage TBSA burn of the dischargedpatientstothoseofthehospitalizedpatients(12.4 ___4.2 years vs 11.1 _+ 3.5 years [mean _+ SD], respectively). Nutritional support of hospitalized patients consisted of a daily nasogastric infusion of milk, usually beginning in the first 24 hours, to achieve between 1500 and 1800 kcal/m 2 body surface area plus an additional 1500 to 2000 kcal/m 2 surface area burned. 6 The milk provided 1.8 mg of elemental calcium per kilocalorie delivered; thus hospitalized patients received a calculated minimum of 2.7 grn of elemental calcium per square meter daily during the acute burn period. All discharged patients were eating regular diets and pursued unrestricted activities; many participated electively in competitive sports. Procedures. Patients in each group underwent measurements of bone mass of the lumbar vertebrae (L1 to L4) by dual-energy x-ray absorptiometry with the QDR 1000W

253

absorptiometer (Hologic Inc., Waltham, Mass.). During the period of the study, the coefficient of variation of the spine phantom in repeated measurements was 0.31%. The bone mass measurements were transformed into z scores, with the use of data on 219 children and adolescents studied by Southard et al.s Heights and weights were determined in all discharged patients at the time of their bone mass measurements. Heights in the hospitalized patients were obtained at admission, and weights were obtained at the time of bone mass measurement. From the 22 hospitalized patients, we obtained 24-hour urine collections for the determination of calcium and phosphate concentrations (n = 12), and blood for determination of serum levels of total calcium, phosphorus, intact parathyroid hormone (n = 22), and serum osteocalcin and type I collagen telopeptide (n = 21). From 12 of the 30 discharged severely burned patients, we obtained blood for serum levels of intact parathyroid hormone, osteocalcin, and type I collagen telopeptide. Calcium and phosphorus were analyzed colorimctrically on an Ektachrome 700 analyzer (Eastman Kodak Co., Rochester, N.Y.). Serum intact parathyroid hormone, 9 osteocalcin, 1° and type I collagen telopeptide 11 were measured as previously described: Additionally, the moderately burned and severely burned patients discharged from the hospital, and their parents, were asked whether the patients had had fractures after release from the initial hospitalizationfor burn, the location of such fractures, and the circumstances under which the fractures occurred. This study was approved by the institutional review board of the University of Texas Medical Branch, and the form for informed consent was reviewed and approved by the Shriners Hospitals for Crippled Children, Tampa, Fla. RESULTS In our study, the age at which the discharged patients acquired their burn did not differ in the moderately and the severely burned groups (Table I). However, the hospitalized severely burned patients were significantly older at injury (p <0.002). All patients were of comparable age when studied (Table I). Moderately burned patients underwent bone mass determination a mean ( _+SD) of 5.8 + 4.4 years (range, 0.3 to 16.2 years) after the injury; discharged, severelyburned patients underwent bone mass determination 5.5 + 3.7 years (range, 0.4 to 13 years) after the injury. Hospitalized severely burned patients underwent bone mass determination 37 _+ 14 days (range, 18 to 59 days) after the burn. The heights and weights of the three groups were not significantly different. The percentage of patients with moderate burns (31%; 5/16) and the percentage of those with severe burns (60%; 31/52) who had a z score for bone mass _<- 1 were increased (p <0.005) (Figure). In the severely burned group, 27% (14/ 52) of the patients had a bone mass z score ~<-2, which differed significantly from the normal distribution (p <0.005).

254

Klein et al.

The Journal of Pediatrics February 1995

Table I. Patient characteristics Patients (No.) M

Age at burn (yr)

Age at study (yr)

Weight* (% ile for age)

Heightt (% ile for age)

F

M

F

M

F

M

F

M

F

Moderate burn (<40%) Discharged 9

7

5.4 (_+4.0)

4.8 (_+3.7)

11.1 (-+4.2)

10.5 (-+4.4)

60 (_+31)

66 (+20)

46 (+34)

53 (_+31)

Severe burn (>40%) Hospitalized 14

8

11.4~ (+3.4) 6.9 (--5.4)

10.8:~ (_+4.1) 7.1 (+4.7)

11.4 (--_3.4) 12.9 (_+4.8)

10.8 (_+4.1) 12.0 (+4.4)

64 (_+29) 39 (--30)

76 (+29) 47 (_+33)

52 (-+35) 32 (-+22)

62 (_+34) 39 (_+29)

Discharged

15

15

All valuesexcept number of patients are givenas mean (+_SD). *Obtained at the time of bone mass determination. tObtained at the time of bone mass determinationin discharged patients and at the time of admissionin hospitalizedpatients. :~p <0.02 versus moderatelyburned and discharged severelyburned groups. Table II. Serum levels of osteocalcin and type I collagen telopeptide in severely burned hospitalized and discharged patients Hospitalized (n = 21) Osteocalcin (tzg/L) Mean 3.1 ___2.2* Range 0.5-7.6 Normal (for age) <12 yr 10-25 >12 yr 2-8 Type I collagen telopeptide 0zg/L) Mean 82.2 -- 43.1 Range 23.9-186.5 Normal (all ages) <30

Discharged (n = 12) 6.3 +_ 2.9 0.8-12.3

18.4 + 9.0t 9.7-44.1

*Mean ± SD. tP < 0.05 compared with hospitalizedpatients. We performed linear regression analysis on the bone mass z score as a function of the percentage of TBSA burned and determined that when the moderate and severe groups were combined, there was a weak but significant correlation (r = -0.3; p = 0.001). We did not find a relation between the time after burn injury and the bone mass z score. There had been no fractures in the moderate burn group. Ten patients (3 girls, 7 boys) in the discharged severe burn group had had fractures; 8 fractures were appendicular (of a limb or digit) and 2 were axial (1 of the hip, 1 of the ribs). All but one of the fractures were from trauma incurred as a result of age-appropriate physical activity; one ankle fracture was pathologic. The extrapolated fracture incidence in the discharged severely burned group was 76.3/ 1000 boys per year and 40.8/1000 girls per year. Incidence of fracture for age-matched healthy children are 41 to 43/ 1000 boys per year and 27 to 30/1000 girls per year, 12, 13 which suggests that the incidence of fractures in this severely burned group may be higher than normal, although the number of patients was too small to perform statistical

analysis in comparison with the large population studies on which the normative data are based. Hospitalized severely burned patients (n = 12) had daily urinary calcium excretion of 50.4 + 30.4 mmol (202 + 122 mg) (range, 11.4 to 118.3 mmol [46 to 474 mg]) or, expressed on a body mass basis, 1.22 _+ 0.81 mmol/kg (4.89 + 3.24 mg/kg) (range, 0.19 to 2.77 mmol/kg [0.77 to 11.10 rag/ kg])) 4 However, the extent of the hypercalciuria correlated neither with bone mass z scores (r = -0.2) nor with the percentage of TBSA burned (r = 0.42). Daily urinary phosphate excretion ranged from 3.2 to 77.5 mmol (0.1 to 2.4 gm). Serum total calcium concentrations in the hospitalized, severely burned group were low on admission (range, 1.8 to 2.1 mmol [7.2 to 8.6 mg/dl]), and rose to the normal range (2.1 to 2.4 mmol [8.5 to 9.6 mg/dl]), by 1 week later. Serum phosphorus and intact parathyroid hormone concentrations remained normal throughout the burn period. Serum osteocalcin concentrations were low for age in 86% (18/21) of the patients, and type I collagen telopeptide concentrations were elevated in 90% (19/21) of the patients (Table II). In the discharged severely burned group, serum osteocalcin levels were reduced for age in 50% (6/12) and normal in the others. Serum type I collagen telopeptide concentrations were normal in 92% (11 / 12) of the patients. Serum intact parathyroid hormone concentrations were normal in all 12 patients. DISCUSSION Our study demonstrates that in children who sustain a severe thermal burn injury, acute loss of bone mass occurs and is maintained. The pathogenesis remains incompletely understood, but our study has added the information that bone formation, as assessed indirectly by serum osteocalcin levels, may be reduced from normal, and bone resorption as assessed indirectly by serum type I collagen telopeptide and urinary calcium excretion may be increased. This disorder of bone homeostasis may increase the susceptibility of burned children to appendicular or axial fractures long after the acute thermal injury. Our

The Journal o f Pediatrics Volume 126, Number 2

Klein et al.

255

61

5

4 a.

3 E z

-5

, F

-4

-3

A

-2

-1

0

1

2

,,

3

4

I 5

B M D z-Score 16 14-

.~12 C

~

-

-68 z~

422 0

-5 -4

B

-3

-2 -1

0

1

2

I

3

I

4

I

5

BMD z-Score

Figu re. A, Distribution of bone mineral density (BMD) z scores in relation to number of patients in the moderately burned group. B, Distribution of bone mass (BMD) z scores in relation to number of patients in the severely burned group.

data on bone turnover are consistent with previously published findings in adults with comparable thermal injury.2 A variety of factors may contribute to the acute bone loss detected in our patients, although roles played by these factors have not been investigated in depth. Immobilization after burn injury may contribute to an acute reduction in bone formation and a reduction in bone mass. Thus Arnaud et al. 15 found a reduction in bone formation in normal young adults at bed rest for only a week, and LeBlanc et al. 16reported that young adults at bed rest for 17 weeks lost up to 10% of their bone mass. Because of the necessity of confining a patient to bed after severe burn injury, it is not possible to determine whether reduced bone formation can occur in patients in the absence of immobilization. Increased production of endogenous corticosteroids may contribute to reduced bone formation, 17 although the rapidity and the extent to which they may do so are currently unknown. Corticosteroids may also contribute to increased bone resorption by decreasing intestinal calcium absorption and thus producing secondary hyperparathyroidism, l 7 a finding that we did not observe in our patients. Moreover, corticosteroids may in-

crease bone resorption by stimulating production of bone-resorbing cytokines, such as interleukin-6.18 A l u m i n u m loading is another well-described cause of reduced bone formation 19 and has been reported to occur in thermal injury. 2 However, we found no evidence in our previous work 2° that aluminum-induced bone disease persists after withdrawal of aluminum-contaminated parenteral solutions, such as albumin or calcium gluconate. 21 Furthermore, neither the rapidity with which aluminum can reduce bone formation nor the quantity of aluminum needed to produce those changes is clear. Thus the reduction in bone formation in adults treated with total parenteral nutrition delivering an aluminum load of 60 gg/kg per day 22 occurred by approximately 8 weeks after initiation of total parenteral nutrition. 23 This quantity was about tenfold greater than that calculated to be received by adult burn patients. 2 The origin of hypercalciuria in the acute postburn period in our patients is not clear. Hypercalciuria may simply be a manifestation of increased bone resorption as reflected in the elevated serum telopeptide levels. However, the large amounts

256

Klein et al.

of dietary protein 2.26 received by our patients, as well as the high intake of dietary calcium, w may also have contributed to the pathogenesis of hypercalciuria. Unfortunately, we have no data on the duration of the hypercalciuria, which might shed more light on its role in the sustained reduction of bone mass. Other factors that may contribute to the bone disease inelude altered body biomechanies produced by soft tissue damage caused by the burn injury. The role of specific surgical techniques in contributing to the bone loss may also need to be investigated. There was no evidence of child abuse in any of the patients participating in the study. Acute malnutrition was not present in any of the severely burned patients, as judged by weight for age, so it is unlikely that malnutrition contributed significantly to bone loss. Furthermore, there is no delay in puberty in burned children that could potentially cause misinterpretation of the bone mass z score. 27 Our cross-sectional data showing that the acute osteopenia is maintained indicates that children do not completely recover from the acute insult and that those who were entering the period of peak bone mass accumulation may develop lifelong osteoporosis. Thus the burn-induced bone loss may be a forerunner of early-onset osteoporosis in later life. A longitudinal study is needed to test this hypothesis. Cross-sectional studies are of limited value in the provision of data on the rate and direction of bone mass alteration during acute thermal injury and long-term recuperation. However, we have demonstrated a significant problem in such patients with significant clinical consequences. Further studies will address intervention to prevent the acute bone loss, and longitudinal study of other patients with burn-associated osteoporosis will provide insights into the pathophysiologic mechanism. REFERENCES

1. Rutan RL, Herndon DN. Growth delay in postburn pediatric patients. Arch Surg 1990;125:392-5. 2. Klein GL, Herndon DN, Rutan TC, et al. Bone disease in burn patients. J Bone Miner Res 1993;8:337-45. 3. Kupper F J, Deitch EA, Biker CC, Wong W. The human burn wound as primary source of interleukin-1 activity. Surgery 1986;100:409-15. 4. Guo Y, Dickerson C, Chrest F J, et al. Increased levels of circulating interleukin-6 in burn patients. Clin Immunol Immunopathol 1990;54:361-71. 5. IshimiY, Miyaurh C, Jin CH, et al. IL-6 is produced by osteoblasts and induces bone resorption. J Immunol 1990; 145:3297-303. 6. Herndon DN, Rutan RL, Alison WE Jr, Cox CC Jr. Management of burn injuries. In: Eichelberger MR, ed. Pediatric trauma prevention acute care, rehabilitation. St Louis: Mosby-Year Book, 1993:568-90. 7. Rutan TC, Herndon DN, Van Osten T, Abston S. Metabolic rate alterations in early excision and grafting versus conservative treatment. J Trauma 1986;26:140-2.

The Journal of Pediatrics February 1995

8. Southard RN, Morris JD, Mahan JD, et al. Bone mass in healthy children: measurements with quantitative DXA. Radiology 1991;179:735-8. 9. Gidding SS, Minciotti A, Langman CB. Unmasking of hyperparathyroidism in familial partial DiGeorge syndrome by EDTA challenge. N Engl J Med 1988;319:1589-91. 10. Reed A, Haugen M, Pachman LM, Langman CB. Abnormalities of serum osteocalcin in children with chronic rheumatic diseases. J PEDIATR 1990;116:574-80. 11. Eriksen EF, Charles P, Melsen F, Mosekide L, Risteli L, Risteli J. Serum markers of type I collagen formation and degradation in metabolic bone disease: correlation with bone histomorphometry. J Bone Miner Res 1993;8:127-32. 12. Holbrook TL, Grazier K, Kelly JL, Stauffer RN. The frequency of occurrence, impact and cost of selected musculoskeleta1 conditions in the United States. Chicago: American Academy of Orthopedic Surgeons, 1984:73-135. 13. Worlick P, Stower M. Fracture patterns in Nottingham children. J Pediatr Orthop 1986;6:656-60. 14. Stapleton FB, Roy S, Noe HN, Jerkins G. Hypercalciuria in children with bematuria. N Engl J Med 1984;310:1345-8. 15. Arnaud S, Sherrard D J, Maloney N, et al. Effects of 1-week head-down tilt bed rest on bone formation and the calcium endocrine system. Aviat Space Environ Med 1992;63:14-20. 16. LeBlanc AD, Schneider VS, Evans H J, Engelbretson DA, Krebs JM. Bone mineral loss and recovery after 17 weeks of bed rest. J Bone Miner Res 1990;5:843-50. 17. Hahn TJ. Steroid and drug-induced osteopenia. In: Favus M J, ed. Primer on the metabolic bone diseases and disorders of mineral metabolism. New York: Raven Press, 1993:250-5. 18. Mastorakos G, Chrousos GP, Weber JS. Recominant interleukin-6 activates the hypothalamic-pituitary-adrenal axis in humans. J Clin Endocrinol Metab 1993;77:1690-4. 19. Klein GL, Coburn jW. Parenteral nutrition: effect on bone and mineral homeostasis. Annu Rev Nutr 1991;11:93-119. 20. Vargas JH, Klein GL, Ament ME, et al. Metabolic bone disease of total parenteral nutrition, course after changing from casein to amino acids in parenteraI solutions with reduced aluminum content. Am J Clin Nutr 1988;48:1070-8. 21. Sedman AB, Klein GL, Merritt R J, et al. Evidence of aluminum loading in infants receiving intravenous therapy. N Engl J Med 1985;312:1337-43. 22. Klein GL, Alfrey AC, Shike M, Sherrard DJ. Parenteral drug products containing aluminum as an ingredient or a contaminant: response to FDA notice of intent. Am J Clin Nutr 1991; 53:399-402. 23. Klein GL, Targoff CM, Ament ME, et al. Bone disease in parenteral nutrition. Lancet 1980;2:1041-4. 24. Margen S, Chu JY, Kaufmann NA, Calloway DH. Studies in calcium metabolism 1: the calciuretic effect of dietary protein. Am J Clin Nutr 1974;27:584-9. 25. American Academy of Pediatrics Committee on Nutrition. Pediatric nutrition handbook. Elk Grove Village, Illinois: American Academy of Pediatrics, 1985:354. 26. Neer RM. Calcium and inorganic phosphate homeostasis. In: De Groot L, ed. Endocrinology. Philadelphia: Saunders, 1989:927-53. 27. Rutan RL, Moore M, Alvarado MI, Blakeney PE, Herndon DN. Growth and development following thermal injury [Abstract]. Proceedings of the 3rd Sino-Ameriean Conference on Burns and Trauma, Canton, China, 1993, p. 171.