longitudinal growth in hiv-negative boys with haemophilia

4
ORIGINAL PAPER Jan Lebl Æ Jutta Falger Æ Thomas Zidek Æ Christoph Male Æ Vladimir Komrska Herwig Frisch Longitudinal growth in HIV-negative boys with haemophilia Received: 5 January 1999 / Accepted: 16 December 1999 Abstract It has been shown that HIV-positive haemophilic children develop growth retardation. As not only the HIV infection but also other disease-related factors might compromise growth in these children, growth data were analysed in a longitudinal cross- sectional manner in 84 HIV-negative haemophilic patients from two university clinics. A total of 2–24 height and weight measurements (median 6) were recorded in each patient resulting in 683 single values collected between 1977–1995. Height SDS of all haemophilic boys was )0.31 2.13 (mean SD, NS versus 0) and body mass index SDS was 0.21 3.49 (mean SD, NS versus 0) at first measurement and remained unchanged throughout the observation period. Neither height nor body mass index diered with respect to the severity of haemophilia (mild/moderate/severe) or the study centre (Vienna/ Prague). Conclusion Growth in HIV-negative patients with haemophilia is not aected in spite of the immunological abnormalities attributed to the substitution therapy or the bleeding episodes in the joints with the potential eect on the growth plate. Key words Haemophilia Æ HIV-negative patients Æ Children Æ Growth Æ Body mass index Abbreviation BMI body mass index Introduction Our knowledge about spontaneous growth in boys with haemophilia is still limited. In the late 1980s, growth was repeatedly studied in HIV-positive haemophilic patients. Poor growth was found in children who were HIV se- ropositive even before the onset of clinical symptoms of AIDS [8, 9]. Subsequently, growth failure was recogn- ised to be a good predictor for progression into clinical AIDS [2]. The extensive Haemophilia Growth and Development Study showed that growth velocity is reduced and pubertal maturation retarded in HIV- positive haemophilic patients compared to HIV-negative haemophilic controls [6]; however, a recent re-analysis in a subgroup of the same cohort of patients [15] proved an abnormal growth and/or pubertal development even in 31 of 126 HIV-negative subjects. These findings promote interest in more detailed studies in HIV-negative boys with haemophilia. Not only the HIV infection but also other factors might compromise the spontaneous growth in these children. The aim of the study was to determine whether severe haemophilia per se and/or its therapy might aect growth in a large cohort of patients. We collected longterm growth data in haemophilic patients from Eur J Pediatr (2000) 159: 575–578 Ó Springer-Verlag 2000 J. Lebl Department of Paediatrics, 3rd Faculty of Medicine, Charles University, Prague, Czech Republic J. Falger Æ C. Male Æ H. Frisch (&) Department of Paediatrics, University of Vienna, Waehringer Guertel 18–20, 1090 Vienna, Austria T. Zidek Department of Epidemiology, Institute of Tumour Biology and Cancer Research, University of Vienna, Vienna, Austria V. Komrska 2nd Department of Paediatrics, 2nd Faculty of Medicine, Charles University, Prague, Czech Republic

Upload: herwig

Post on 25-Aug-2016

212 views

Category:

Documents


0 download

TRANSCRIPT

ORIGINAL PAPER

Jan Lebl á Jutta Falger á Thomas Zidek á Christoph Male á Vladimir KomrskaHerwig Frisch

Longitudinal growth in HIV-negative boyswith haemophilia

Received: 5 January 1999 /Accepted: 16 December 1999

Abstract It has been shown that HIV-positive haemophilic children develop growthretardation. As not only the HIV infection but also other disease-related factors mightcompromise growth in these children, growth data were analysed in a longitudinal cross-sectional manner in 84 HIV-negative haemophilic patients from two university clinics. Atotal of 2±24 height and weight measurements (median 6) were recorded in each patientresulting in 683 single values collected between 1977±1995. Height SDS of all haemophilicboys was )0.31 � 2.13 (mean � SD, NS versus 0) and body mass index SDS was0.21 � 3.49 (mean SD, NS versus 0) at ®rst measurement and remained unchangedthroughout the observation period. Neither height nor body mass index di�ered withrespect to the severity of haemophilia (mild/moderate/severe) or the study centre (Vienna/Prague).

Conclusion Growth in HIV-negative patients with haemophilia is not a�ected in spite ofthe immunological abnormalities attributed to the substitution therapy or the bleedingepisodes in the joints with the potential e�ect on the growth plate.

Key words Haemophilia á HIV-negative patients á Children á Growth á Body mass index

Abbreviation BMI body mass index

Introduction

Our knowledge about spontaneous growth in boys withhaemophilia is still limited. In the late 1980s, growth wasrepeatedly studied in HIV-positive haemophilic patients.Poor growth was found in children who were HIV se-ropositive even before the onset of clinical symptoms ofAIDS [8, 9]. Subsequently, growth failure was recogn-ised to be a good predictor for progression into clinicalAIDS [2]. The extensive Haemophilia Growth andDevelopment Study showed that growth velocity isreduced and pubertal maturation retarded in HIV-

positive haemophilic patients compared to HIV-negativehaemophilic controls [6]; however, a recent re-analysis ina subgroup of the same cohort of patients [15] proved anabnormal growth and/or pubertal development even in31 of 126 HIV-negative subjects. These ®ndings promoteinterest in more detailed studies in HIV-negative boyswith haemophilia. Not only the HIV infection but alsoother factors might compromise the spontaneous growthin these children.

The aim of the study was to determine whether severehaemophilia per se and/or its therapy might a�ectgrowth in a large cohort of patients. We collectedlongterm growth data in haemophilic patients from

Eur J Pediatr (2000) 159: 575±578 Ó Springer-Verlag 2000

J. LeblDepartment of Paediatrics, 3rd Faculty of Medicine,Charles University, Prague, Czech Republic

J. Falger á C. Male á H. Frisch (&)Department of Paediatrics, University of Vienna,Waehringer Guertel 18±20, 1090 Vienna, Austria

T. ZidekDepartment of Epidemiology, Institute of Tumour Biologyand Cancer Research, University of Vienna, Vienna, Austria

V. Komrska2nd Department of Paediatrics, 2nd Faculty of Medicine,Charles University, Prague, Czech Republic

University departments from two neighbouring central-European countries with low prevalence of HIV infec-tion. The data were compared to age-speci®c growthstandards allowing for the type of haemophilia and theseverity of the disease.

Subjects and methods

Growth data of 84 boys with haemophilia, born between 1975 and1994, were analysed. Of these patients, 41 were under the care ofthe Department of Paediatrics, University Hospital Vienna, Aus-tria and 43 were in the 2nd Department of Paediatrics, UniversityHospital Prague-Motol, Czech Republic. All of them were HIV-1and HIV-2 negative, as proven by an enzyme immunoassaythroughout the study. A total of 78 of the subjects studied su�eredfrom haemophilia A, the remaining 6 from haemophilia B.

According to the factor VIII or IX level, haemophilia wasclassi®ed as mild (F VIII or IX >5%) in 24 of the subjects (28.6%),as medium (F VIII or IX 2%±5%) in 16 of the subjects (19.0%)and as severe (F VIII or IX <2%) in 44 of the subjects (52.4%),respectively. The clinical status of patients was classi®ed accordingto Gilbert [7]. The mean score of the entire group of boys was1.15 � 0.24 (mean � SEM) and may be regarded as good.

Growth data were evaluated in a longitudinal/cross-sectionalmanner on 2±24 occasions in each subject (median 6). Only datathat were at least 6 months apart were included. A total number of683 height and weight measurements was recorded between 1977and 1995 in the age range 0.2±17.6 years. Height was measuredusing a wall-mounted stadiometer by a trained person. Theparental heights were available for analysis in 47 patients. Paternal

heights (178.7 � 6.2 cm, mean � SD) and maternal heights(165.5 � 5.6 cm, mean � SD) were no di�erent from the popu-lation standards.

Population standards

The Zurich longitudinal growth study was used as a populationstandard for height and for body mass index (BMI) [13]. Data forBMI from the same study were kindly supplied by A. Prader,ZuÈ rich. These data are based on measurements of a populationgeographically and ethnically comparable to the study population.The timespan of collection of reference data (year of publication1989) corresponds quite fairly with the observation period in thestudy population (1977±1995) which makes any in¯uence of secularchanges unlikely.

Statistical evaluation

The standardisations of height and of BMI were done using thestandards of Prader et al. [13]. After standardisation, a linearregression for every patient's data was performed and the interceptand coe�cients were tested using the Wilcoxon Ranksum test (nullhypothesis median � 0). This procedure was provided to rule out agreater impact of patients with more available data (which couldlead to a bias towards patients with more frequent measurements

Fig. 1 a Individual values of standardised height of patients withhaemophilia. b Height percentiles of boys with mild, moderate orsevere haemophilia based on the analysis of individual regression linesof standardised height, on the background of a normal growth chart

576

and/or a longer observation period). Additionally a descriptiveanalysis of percentile curves was performed using GROSTAT [14].

For statistical procedures, the SUPSMU function of the SPLUSpackage was used [4]. To diminish the statistical in¯uence of extremevalues, non parametric tests were applied. All tests were done 2-sided. P values <0.05 were considered statistically signi®cant.

Results

Body height

The height SDS in the entire group of haemophilicsubjects at the ®rst measurement did not signi®cantlydi�er from zero and remained unchanged during the

observation period. The intercept of the regression linewas )0.31 � 2.13 (mean � SD, NS) and the slope0.05 � 0.28 (NS) (Fig. la, b). The standardised growthdata were no di�erent either between patients fromboth study centers or between patients with di�erentdisease severity (Table 1).

Body mass index

Also the BMI SDS in the whole group of patients at the®rst measurement did not di�er from zero and remainedunchanged throughout the observation period. The in-tercept of the regression line was 0.21 � 3.49(mean � SD, NS) and the slope 0.05 � 0.54 (NS)(Fig. 2a, b). The standardised values of BMI did notdi�er either according to the study centre or to the se-verity of the disease (Table 1).

Fig. 2 a Individual values of standardised BMI of patients withhaemophilia. b Percentiles of BMI of boys with mild, moderate orsevere haemophilia based on the analysis of individual regression linesof standardised height, on the background of a normal growth chart

Table 1 Intercept and slope(mean � SD) of individualregression lines of height-SDSand BMI-SDS. There was nodi�erence in results between thetwo study-centres or the di�er-ent degrees of severity

Body height BMI

Intercept mean (SD) Slope mean (SD) Intercept mean (SD) Slope mean (SD)

Vienna )0.62 (2.42) 0.07 (0.35) 0.12 (3.08) 0.06 (0.63)Prague )0.02 (1.79) 0.04 (0.21) 0.30 (3.88) 0.04 (0.44)Mild )0.81 (2.80) 0.13 (0.41) 0.12 (3.14) 0.05 (0.54)Medium 0.02 (2.61) 0.00 (0.15) 0.80 (3.05) 0.03 (0.43)Severe )0.17 (1.41) 0.03 (0.23) 0.05 (3.85) 0.05 (0.58)

577

Discussion

Many chronic diseases may a�ect statural growth duringchildhood. In some of them, the mechanism leading togrowth retardation is strictly hormonal (growthhormone de®ciency, hypothyroidism, glucocorticoidexcess), in others nutritional (poor nutrition, malab-sorption), immunological (juvenile chronic arthritis,Crohn disease) or other factors may play a role.

Immunological disorders have been described inhaemophilic patients even if they were HIV-negative.These include abnormalities in circulating lymphocytesubsets [3], decreased production of interleukin-2 [10] orof interferon-gamma [12], B-cell dysfunction [11], im-paired monocyte chemotaxis and phagocytic function[17] or de®cient skin test responses [1]. Findings sugges-tive of immunosupression might be due to the action oftransforming growth factor-b which has been found as acontaminant in many of the factor VIII concentrates [18,19]. It is well documented that cytokines may interferewith insulin-like growth factor-1 production [20] and achronic immunopathological state even unrelated to HIVinfection may lead to growth retardation [5].

Another reason for growth abnormalities in haemo-philic patients might be their bleeding episodes in thejoints. In immature articulation, subsequent synovitisdirectly a�ects the growth plate [16] with an impact onthe long-term growth potential of long bones. We ob-served individual HIV-negative haemophilic boys withgrowth retardation. No explanation could be found in atleast one boy with severe haemophilia A who su�eredfrom poor growth ()3.2 SD at 5 years of age) and froma proportional bone age delay in spite of normal hor-monal status and absence of any systemic disease. Hisgrowth retardation could be attributed to an extremeform of constitutional delay of growth or, as hypothes-ised, high cumulative dose of factor VIII concentratesubstitution or to the consequences of joint bleeding.Obviously none of these theories plays a role in HIV-negative haemophilic children as we found that theseboys grow normally with no relation to the severity orthe type of the disease and thus independent of theircumulative dose of factor VIII concentrate substitution.This would be important information for haemophilicchildren and a challenge for their physicians to searchfor non-haemophilia related conditions in those patientswith growth retardation.

The growth of haemophilic boys did not di�er fromthat of the control population which means that normalheight standards can be used for the care of these pa-tients. Fortunately, disease-speci®c standards are notnecessary.

References

1. Brettler DB, Forsberg AD, Brewster F, Sullivan JL, Levine PH(1986) Delayed cutaneous hyper-sensitivity reactions in hemo-

philic subjects treated with factor concentrate. Am J Med 81:607±611

2. Brettler DB, Forsberg A, Bolivar E, Brewster F, Sullivan J(1990) Growth failure as a prognostic indicator for progressionto acquired immunode®ciency syndrome in children withhemophilia. J Pediatr 117: 584±588

3. Carr R, Veitch SE, Edmond E, Peutherer JF, Prescott RJ,Steel CM, Ludlam CA (1984) Abnormalities of circulatinglymphocyte subsets in haemophiliacs in an AIDS-free popula-tion. Lancet 1 : 1431±1434

4. Chambers JM, Hastie TJ (1991) Statistical models. In: Wads-worth S, Brooks/Cole Advanced books and software, Paci®cGrove, Calif

5. De Benedetti F, Alonzi T, Moretta A, Lazzaro D, Costa P, PoliV, Martini A, Ciliberto G, Fattori E (1997) Interleukin 6 causesgrowth impairment in transgenic mice through a decrease ininsulin-like growth factor-1. A model for stunted growth inchildren with chronic in¯ammation. J Clin Invest 99: 643±650

6. Gertner JM, Kaufman FR, Don®eld SM, Sleeper LA, ShapiroAD, Howard C, Gomperts ED, Hilgartner MW (1994) Delayedsomatic growth and pubertal development in human immun-ode®ciency virus-infected hemophiliac boys: hemophilia growthand development study. J Pediatr 124: 896±902

7. Gilbert MS (1993) Prophylaxis: musculoskeletal evaluation.Semin Hematol 30[Suppl 2]: 3±6

8. Jason J, Gomperts E, Lawrence DN, Holman RC, BouhasinJD, Miller R, Evatt BL (1989) HIV and hemophilic children'sgrowth. J Acquir Immune De®c Syndr 2: 277±282

9. Kaufman FIR, Gomperts ED (1989) Growth failure in boyswith hemophilia and HIV infection. Am J Pediatr HematolOncol 11: 292±294

10. Madhok R, Gracie JA, Smith J, Lowe GD, Forbes CD (1990)Capacity to produce interleukin-2 is impaired in haemophilia inthe absence and presence of HIV-1 infection. Br J Haematol 76:70±74

11. Madhok R, Gracie A Forbes CD, Lowe GD (1991) B celldysfunction in haemophilia in the absence and presence ofHIVA infection. Thromb Haemost 65: 7±10

12. Newton-Nash DK, Tollerud D, Guevarra L, Gill JC (1996)Interferon-gamma secretion defects in haemophilia A patientsreceiving highly puri®ed plasma-derived or recombinant factorVIII. Brit J Haematol 95: 554±560

13. Prader A, Largo RH, Molinari L, Issler C (1989) Physicalgrowth of Swiss children from birth to 20 years of age. HeIvPaediatr Acta 52[Suppl]: 1±125

14. Rasbash J, Pan H, Goldstein H (1989) Grostat: A program forestimating age related centiles using piecewise polynomials.WHO, Geneva

15. Ratner-Kaufman F, Gertner JM, Sleeper LA, Don®eld SM(1997) Growth hormone secretion in HIV-positive versus HIV-negative hemophilic males with abnormal growth and pubertaldevelopment. The hemophilia growth and development study.J Acquir Immune De®c Syndr Hum Retrovirol 15: 137±144

16. Rodriguez-Merchan EC (1996) E�ects of hemophilia on artic-ulations of children and adults. Clin Orthop 328: 7±13

17. Roy G, Pardo A, Leyva-Cobian F (1988) Phenotypic andfunctional abnormalities in monocytes from patients withhaemophilia A treated with FVIII concentrates. Acta Haematol79: 26±32

18. Wadhwa M, Dilger P, Tubbs J, Mire-Sluis A, Barrowcli�e T,Thorpe R (1994) Identi®cation of transforming growth factor-bas a contaminant in factor VIII concentrates: A possible linkwith immunosuppressive e�ects in hemophiliacs. Blood 84:2021±2030

19. Wadhwa M, Barrowcli�e TW, Mire-Sluis AR, Thorpe R (1995)Factor VIII concentrates and the immune system ± laboratoryinvestigations. Blood Coagul Fibrinolysis 6[Suppl 2]: S65±S79

20. Wolf M, Bohm S, Brand M, Kreymann G (1996) Proin¯am-matory cytokines interleukin 1 beta and tumor necrosis factoralpha inhibit growth hormone stimulation of insulin-likegrowth factor I synthesis and growth hormone receptor mRNAlevels in cultured rat liver cells. Eur J Endocrinol 135: 729±737

578