Pathology in Children of HIV Women - MedIND

Original Article

Pathology in Children of HIV Women A.P. Nso, C García-Bermejo, B. Larru1, R. Madero, M.A. Muñoz Fernández1 and M.I. de José Servicio Inmuno-Pediatría; Hospital Universitario “La Paz”, 1 Laboratorio de Inmuno-Biología Molecular, Hospital General Universitario Gregorio Marañón, Madrid, Spain

ABSTRACT Objective. To assess the frequency of perinatal pathology in children exposed to antiretrovirals in perinatal period. Methods. Retrospective observational cohort study. Data collected among uninfected children born to HIV-infected women followed up from 1994 to 2006 in a tertiary Hospital. 220 uninfected children were studied. Factors studied included maternal, obstetrical and pediatric variables. Results. The most common disorder found among children exposed to antiretroviral drugs was anemia (84%); 6,4% of children had neutropenia and more than 24% had thrombocytosis, a finding never described before. Prematurity (24%) and low birth weight (23.6%) rates were high. Several congenital malformations were found: Poland syndrome, angiomas, hypospadias, Pierre-Robin sequence, trisomy 8, craniostosis and others. Long-term follow-up revealed neurological, cardiological and ophthalmological pathologies. Conclusion. Some pathologies are frequent among children exposed to antiretroviral agents during perinatal life. It is crucial to carry out long-term studies to assess the safety of this therapy. [Indian J Pediatr 2009; 76 (11) : 1125-1130] E-mail: [email protected]

Key words: AIDS; Pregnancy; Antiretroviral treatment; Adverse effects; Perinatal pathology

The number of women living with HIV has increased everywhere in the last two years, until the point that, nowadays, HIV infected women represent, in several parts of the world, more than half of the infected adult population. 1 Numerous studies have demonstrated the effectiveness of the antiretroviral therapy in reducing the transmission of HIV from mother to child.2 Those studies confer to the programmed cesarean section an important role in the prevention of this transmission.3 In 1997 the use of high activity antiretroviral therapy (HAART) during pregnancy started. At the present time, this regimen is included in the routine prenatal managing of HIV infection, with consequent reduction of the cases of pediatric AIDS acquired perinatally (below 1% in countries with treatment available and good sanitary control).4 The prevention of HIV infection entails, therefore, the prenatal and neonatal exposure to one or more drugs Correspondence and Reprint requests : Dr. Ana Pilar Nso Roca, Servicio de Enfermedades Infecciosas. Hospital Infantil La Paz. Paseo de la Castellana 261; 28046 – Madrid, Spain. [DOI-10.1007/s12098-009-0225-8] [Received June 20, 2008 ; Accepted October 15, 2008]

Indian Journal of Pediatrics, Volume 76—November, 2009

with still unknown toxicity. There are many studies about the pharmacokinetics and the effectiveness of the antiretroviral therapy during pregnancy. 5-7 Nevertheless, questions still exist about the possible short-term and long term risks of that treatment during pregnancy in the exposed children. We consider that the study of the possible adverse effects of a treatment regimen more established nowadays is crucial. MATERIALS AND METHODS We carried out a cohort observational study with retrospective collection of data, after the approval of the project by the Investigation Hospital Commission and the Ethics Committee on Clinical Research. We reviewed charts of uninfected children born to HIVinfected women receiving antiretroviral therapy (ART) that were controlled in our hospital between 1994 and 2006. After obtaining an informed consent, we recorded the following variables: maternal (age, infection route – intravenous, heterosexual or unknown-, co-infection by HCV and HBV, alcohol and tobacco consumption during pregnancy), obstetrical (class of antiretroviral 1125

A.P. Nso et al therapy, treatment period) and pediatric (gestational age, birth weight, HCV and BHV serology, serial hematological studies, congenital malformations, other alterations during the observational period). The children were followed by a consistent team, according to the same protocol. Follow-up visits at our hospital consisted on clinical, laboratory and serological exams at birth, 3 and 6 wk of life and, afterward, at 3, 6, 12 and at 18 mo, when seroreversion was confirmed. Thereafter, clinical and laboratory exams were performed at 24 mo of life and at 5, 7 and 10 yr old. We defined Low birth weight was desined as a birth weight lower than 10 percentile for their gestational age, and high birth weight as a birth weight higher than 95 percentile for the gestational age (according to Lubchenco tables. 8 We defined anemia as a hemoglobin level lower than 10 g/dl at sixth wk of life. We considered neutropenia when neutrophils were < 1500/ µl, thrombocytopenia when platelets were < 150000/µl and thrombocytosis when platelets were > 400000/ µl. Afterward, the statistical analysis was made, keeping the confidentiality of the data recorded. Continuous and categorical data were expressed as median and percentages, respectively. Qualitative data association was analyzed with chi-square test or Fisher exact test according to the data distribution. Quantitative data association was analyzed with Student t test or Mann-Whitney U according to the parametric or non-parametric data distribution. When possible, we estimated a multivariate analysis to determine factors that may be independently associated with each of the more interesting pathologies (drug use, smoking and alcohol consumption during pregnancy). The general population frequency data were obtained from different sources, detailed at tables 1 and 2, the majority of them from the Spanish Dysmorphology and Epidemiology Bulletin. All the statistical test were considered as bilateral and we considered as statistically significant when p value was < 0.05. Statistical analysis was performed using SPSS 15.0 software (SPSS Inc.).

RESULTS

The maternal mean age was 32.5 yr (range 18-42 yr). We observed a change in the transmission route, so that in first half of study period (1994-1999) parenteral transmission was more frequent (52.2%) and in the last years, the main route was the heterosexual one (56%). All the mothers received ART during their pregnancy or delivery. We classified the type of ART in groups according to the number of drugs and the use or not of protease inhibitors. 93% of the mothers were treated with combined therapy. 97.7% of the mothers received ART since the first trimester of their pregnancy. This can account for the high percentage of women receiving combined therapy since the first trimester, regimen adopted after 1996. Data about the type of treatment were incomplete in some mothers, mainly before 1999. The mean age of follow up of the children of these mothers was 3.9 yr (range: 1 mo-10.5 yr). All the children received treatment with Zidovudine since the first 6 hr of life and during the first 6 wk of life and none of them were infected by HIV (only 4 of the 206 children were not followed beyond 18 mo of age and we considered that the serology was negative). All the children underwent blood test at 6 wk but one child that was 4 wk old at the moment of the study. Regarding HBV and HCV infection, no child was HBV positive and only 3 children were HCV positive (rate of HCV vertical transmission in HIV co-infected mothers: 4.8%). Study outcomes We found several disorders in these children. The most frequent were hematological disorders: 84.1% of the children had anemia at 6 wk of life, defined as hemoglobinemia lower than 10 g/dl (mean hemoglobin level of 9.5 g/dl). Anemia was macrocytic (mean corpuscular volume of 102.1 fl) and asymptomatic in most of the cases. Anemia was more frequent in the group of children exposed to combined therapy without protease inhibitors (90.5%), followed by combined therapy with inhibitors of protease (87.5%) and finally by single therapy with Zidovudine (75%). Thrombocytosis was frequent (24.5% of the children), with a mean number of platelets of 695980 /µl. 6.4% of the children presented some type of neutropenia at sixth wk of life: 58.9% light (1000-1500 neutrophils/µl), 38.9% moderate (500-1000 neutrophils/ µl) and 2.2% severe neutropenia (< 500 neutrophils/µl).

Two hundred and twenty (220) mother-child couples were analyzed. Only 14 children (6.4%) were lost in the follow up at a mean age of 10 mo. Therefore, all the results of the study were obtained with the analyses of 206 patients.

There were few data about congenital malformations, defined as alterations that were present at the moment of the birth. In the present study, 28 malformations were found. These malformations are summarized in table 1, where we detail the frequency in general population. This type of alterations was slightly more frequent in the group of treatment that did not

1126


Patient characteristics

Pathology in Children of HIV Women TABLE 1. Congenital Malformations in Our Sample and in General Population Malformation Poland syndrome Angioma Anterior anus Preauricular appendix Cataract Phocomelia Hypospadias Pes Echinovarus Polydactyly Cardiopathy Single umbilical artery Craniostosis Pierre-robin Trisomy 8 Clinodactyly Peculiar face Macrocephaly Williams syndrome

N 1 2 2 1 1 1 2 1 1 2 2 2 1 1 2 3 2 1

SF (%) 0.45 0.91 0.91 0.45 0.45 0.45 0.91 0.45 0.45 0.91 0.91 0.91 0.45 0.45 0.91 1.36 0.91 0.45

PF (%) 0.0019 27 0.065 27 0.0155 27 0.0718 27 0.0203 28 0.0485 27 0.1077 27 0.0327 0.034 27 0.1649 27 0.5- 29 0.0530 0.0130 0.0130 4.6531 NR NR 0.005 30

weight, 0.9% had high birth weight and 23.6% low birth weight. The distribution of the children with low birth weight and the relation with the type of treatmentexposure is represented in fig 1. Prevalence of low birth weight was higher (in some cases even twofold) in the group of treatment without protease inhibitors (33.4%) than with protease inhibitors (21%), but without statistically significance. We did not observe any case of death or development of tumors.

(N: number of cases, SF: sample frequency, PF: population frequency, NR: not reported).

include protease inhibitors (16.7%), but with no significant difference. In long-term clinical and laboratory controls of these children, different alterations were found. These are shown in table 2. Other alterations described in these children were prematurity (gestational age lower than 37 wk) and low birth weight (less than 2.500 grams). In our study, 21.4% of the children were preterm, this represents a frequency significantly higher than one of the general population: 5.2%,9,10 (p < 0.002). The prematurity was more frequent in the group exposed to therapy without protease inhibitors (28.6%) than in the group with protease inhibitors (16.3%), but the differences were not significant. Regarding birth weight, the mean was 2773 gr. (1240-4150 gr). 75.5% of the children had normal birth TABLE 2. Long-term alterations in our sample and in general population Alteration Renal pelvis ectasia Pyloric stenosis Psychomotor retardation Subependymary hemorrhage Nystagmus Visual and auditory deficit Strabismus Febrile seizure Hyperactivity Inguinal hernia Rickets Precocious puberty Umbilical hernia

N 8 1 8 1 1 1 9 1 5 2 4 1 3

(%) 3.64 0.45 3.64 0.45 0.45 0.45 4.09 0.45 2.27 0.91 1.82 0.45 1.36

PF (%) 232 0.333 1.1-2.5 29,34 235 0.0153 36 0.537 429 329 529 3.529 NR 0.03 38 16.66 39

(N: number of cases, SF: sample frequency, PF: population frequency, NR: not reported).


Fig. 1. Birth weight (BW) in the sample in the different treatment groups. (PI: protease inhibitors).

DISCUSSION Since the beginning, questions about the possible risks of the exposure to ART during the perinatal period have been established. Several disorders have been reported11 but the results of the studies are, in some cases, contradictory and the experience in children is limited. After almost 10 years of use of the HAART, we consider necessary an evaluation of its possible adverse effects in our environment. Our study has some limitations due to the difficulty of data collection from charts and the low prevalence of some of the found problems. Nevertheless, we include a high number of patients, and long-term follow-up has been performed, which allows obtaining consistent results. The hematological problems are probably the most frequently described alterations in the literature. 12-14 Secondary effects in the three hematological series have been described: anemia, 2,13 thrombocytopenia15, 16 and neutropenia.17,18 In our sample, the frequency of anemia in the first 6 wk of life is higher than 80%, with no relation with the mother acquisition way of the HIV infection. This frequency is higher than described in others series, around 27%. 2, 4 Like in other studies, 1127

A.P. Nso et al anemia is not severe in most of the cases and it improves with treatment. In our sample, the high frequency of thrombocytosis is a new finding not described before. Previous studies in animals have demonstrated that zidovudine significantly increased platelet production.19 However, thrombocytosis may be multifactorial (iron or vitamin E deficiency, inflammatory diseases), with minor viral infections being one frequent origin in infants. There were no thrombotic complications and no specific therapy was required. Platelet counts alteration and neutropenia, like anemia, are findings without significant clinical outcomes. Therefore, we can conclude that hematological alterations are, clearly, the most frequent disorder in our sample with also the least important clinical consequences. Drug use, smoking and alcohol consumption were not confounders in our study. Different studies about adverse effects of ART show that there are few data about the incidence of congenital malformations in children exposed to these drugs during perinatal life.20 In the present series, the majority of the mothers received ART from the first trimester of the gestation; therefore, most of the studied children were exposed to these drugs since the organogenesis period. This increases the probability that the mentioned findings could be due to side effects of those drugs. But it is very difficult to draw conclusions with our sample size of children and a retrospecitve approach. We found that all the children with malformations were exposed to triple therapy, being similar the proportion of exposed and not exposed to protease inhibitors. Maternal mean age of the children with malformations is 31 years, thus we initially rejected that those malformations were due to advanced maternal age. Although a newborn exposed to HAART during intrauterine life was physically normal at birth, we can not assume that the ART have had no effect.21 For this reason, it is important to make a long-term follow up of these children. In the present study, we have controlled the exposed children for more than 10 years which is one of the longest period performed so far.2,22 We have found more than 10 types of different alterations throughout the years of study. We have found several pathologies to be more prevalent among children exposed to ART during perinatal life than in general population ( Tables 1 and 2). Among others, we emphasize the presence of neurological alterations as nystagmus, psychomotor delay, visual and auditory disorders and hyperactivity. Previous studies 16,23 attribute these neurological manifestations to toxic effects of these drugs in the mitochondria. Nevertheless, they are non-specific alterations and its association with mitochondrial 1128

toxicity is difficult to assess. Moreover, some of these problems may have other roots as strabismus and subependymal hemorrhage that are associated with prematurity or hyperactivity that may have a psychosocial background. Several studies suggest that perinatal exposure to ART affects the gestational age and the birth weight. Despite contradictory results of some works, in the present study we clearly observed a rate of prematurity significantly higher in our children than in general population 9 (p < 0.002). In order to analyze these results, it is necessary to consider that the majority of deliveries of HIV-positive women are by elective cesarean section before week 38 of gestation, sometimes between 36 and 37 weeks. 24 Hence the prophylaxis measures, themselves, make these children become preterm more frequently. To evaluate if the prematurity is due to a probable effect of ART exposure, it would be necessary to have data of the moment and the spontaneity of the labor. These data were not available in all of the patients of our study. Considering that elective cesarean sections are not usually performed before week 36, we can say that the rate of prematurity in our sample, not due to elective cesarean section (below week 36 of gestation), is 15%. This means a prematurity rate 3 times higher in children exposed to HAART during the intrauterine life than in the general population (5%) 9 (p < 0.001). Prematurity can be secondary to HIV infection by itself, but there is no way to determine whether the preterm delivery is due to HIV infection or not. With regard to birth weight, children with birth weights between 1.500 to 2.499 gr are more frequent from mothers treated with ART than in the general population, without relation with the maternal intravenous drug consumption during the pregnancy. Against the results of other studies,25 low birth weight is more frequent in children exposed to combined therapy without protease inhibitors. We have not found statistically significant differences in the group of birth weight lower than 1500 gr, probably because extremely low birth weight is a very exceptional event and we have few cases to be able to find differences. In the present study, like in other previous works,26 we neither observed cases of death nor development of tumors. It is obvious that to go beyond the AIDS epidemic, treatment and prevention efforts must be accelerated simultaneously. The benefits of the use of ART, even during pregnancy, are unquestionable and exceed the probable risks of this therapy. In the last years, problems possibly associated with the exposure to this type of treatment have been described. The results of the present study also indicate probable association of the Indian Journal of Pediatrics, Volume 76—November, 2009

Pathology in Children of HIV Women perinatal exposure to the HAART with the appearance of congenital and long-term pathologies. Initially, these findings could not weaken the advantages of the HAART; nevertheless, they are getting more important at the same time that new ART are being developed. The alterations that we have found are, sometimes, more related to a class of regimen than to another. Longterm studies are needed to confirm the probable adverse effects of each group of these drugs. Even though those effects would not change the rules about the HAART use during the gestation, they can improve the recommendations about the kind of regimen to use. We can conclude that our results show that some pathologies are prevalent among children exposed to ART during perinatal life. It is crucial to carry out longterm studies to assess the safety of this therapy. Contributions: APN had primary responsibility for protocol development, collecting and recording data, and outcome assessment and contributed to the writing of the manuscript. CG and BL contributed to the writing of the manuscript. RM participated in analytic framework for the study. MIJ was responsible for patient screening, and contributed to the writing of the manuscript. MAM supervised the design and execution of the study, the final data analyses, and the writing of the manuscript. All authors read, and the writing of the manuscript.

9.

10.

11.

12.

13.

14.

15.

Conflict of Interest: None Role of Funding Source: None

16. 17.

REFERENCES 1. Joint United Nations Program on HIV/AIDS (UNAIDS) and World Health Organization (WHO). http:// data.unaids.org/pub/EpiReport/2006/2006_EpiUpdate_en.pdf; 2006. 2. Connor EM, Sperling RS, Gelber R, Kiselev P, Scott G, O’Sullivan MJ et al. Reduction of maternal-infant transmission of human immunodeficiency virus type 1 with zidovudine treatment. Pediatric AIDS Clinical Trials Group Protocol 076 Study Group. N Engl J Med 1994;331: 11731180. 3. Elective caesarean-section versus vaginal delivery in prevention of vertical HIV-1 transmission: a randomised clinical trial. The European Mode of Delivery Collaboration. Lancet 1999;353:1035-1039. 4. Thorne C, Newell ML. Prevention of mother-to-child transmission of HIV infection. Curr Opin Infect Dis 2004;17:247-252. 5. Thorne C, Newell ML. Antenatal and neonatal antiretroviral therapy in HIV-infected women and their infants: a review of safety issues. Med Wieku Rozwoj 2003;7:425-436. 6. Alimenti A, Burdge DR, Ogilvie GS, Money DM, Forbes JC. Lactic acidemia in human immunodeficiency virusuninfected infants exposed to perinatal antiretroviral therapy. Pediatr Infect Dis J 2003;22: 782-789. 7. Loubeyre-Unique C, Gautier A, Vauzelle-Gardier C, Champart AM, Bavoux F. [Antiretroviral agents and pregnancy: mitochondrial dysfunction and nucleoside analogs]. Therapie 2001;56:261-266. 8. Lubchenco LO, Hansman C, Dressler M, Boyd E. Intrauterine Growth as Estimated from Liveborn Birth-


18.

19.

20.

21.

22.

23.

24.

25.

Weight Data at 24 to 42 Weeks of Gestation. Pediatrics 1963;32:793-800. Marti C, Pena JM, Bates I, Madero R, de Jose I, Pallardo LF et al. Obstetric and perinatal complications in HIV-infected women. Analysis of a cohort of 167 pregnancies between 1997 and 2003. Acta Obstet Gynecol Scand 2007;86:409-415. De la Torre J, Coll C, Coloma M, Martín J, Padrón E, gonzález N. Pregnancy control in immingrants. www.cfnavarra.es/salud/anales/textos/vol29/sup1/suple5a.html. El Beitune P, Duarte G, Quintana SM, Figueiro-Filho EA, Marcolin AC, Abduch R. Antiretroviral therapy during pregnancy and early neonatal life: consequences for HIVexposed, uninfected children. Braz J Infect Dis 2004;8:140150. Mirochnick M, Stek A, Acevedo M, Keller M, Holland D, Capparelli E et al. Safety and pharmacokinetics of nelfinavir coadministered with zidovudine and lamivudine in infants during the first 6 weeks of life. J Acquir Immune Defic Syndr 2005;39:189-194. Lorenzi P, Spicher VM, Laubereau B, Hirschel B, Kind C, Rudin C et al. Antiretroviral therapies in pregnancy: maternal, fetal and neonatal effects. Swiss HIV Cohort Study, the Swiss Collaborative HIV and Pregnancy Study, and the Swiss Neonatal HIV Study. AIDS 1998;12:F241247. Exposure to antiretroviral therapy in utero or early life: the health of uninfected children born to HIV-infected women. J Acquir Immune Defic Syndr 2003;32:380-387. Adewuyi J, Chitsike I. Haematologic features of the human immunodeficiency virus (HIV) infection in Black children in Harare. Cent Afr J Med 1994;40:333-336. Mofenson LM. Perinatal exposure to zidovudine—benefits and risks. N Engl J Med 2000;343:803-805. Englund JA, Baker CJ, Raskino C, McKinney RE, Petrie B, Fowler MG et al. Zidovudine, didanosine, or both as the initial treatment for symptomatic HIV-infected children. AIDS Clinical Trials Group (ACTG) Study 152 Team. N Engl J Med 1997;336:1704-1712. Krogstad P, Lee S, Johnson G, Stanley K, McNamara J, Moye J et al. Nucleoside-analogue reverse-transcriptase inhibitors plus nevirapine, nelfinavir, or ritonavir for pretreated children infected with human immunodeficiency virus type 1. Clin Infect Dis 2002;34:991-1001. Chow FP, Sutton PA, Hamburger AW. 3'-Azido-3'deoxythymidine ameliorates the thrombocytopenia observed in a murine model of AIDS. Exp Hematol 1990;18:1038-1041. De Santis M, Cavaliere AF, Caruso A, Villa P, Tamburrini E, Cauda R et al. Hemangiomas and other congenital malformations in infants exposed to antiretroviral therapy in utero. JAMA 2004;291:305. Mantovani A, Calamandrei G. Delayed developmental effects following prenatal exposure to drugs. Curr Pharm Des 2001;7:859-880. Fundaro C, Genovese O, Rendeli C, Tamburrini E, Salvaggio E. Myelomeningocele in a child with intrauterine exposure to efavirenz. AIDS 2002;16:299-300. Blanche S, Tardieu M, Rustin P, Slama A, Barret B, Firtion G et al. Persistent mitochondrial dysfunction and perinatal exposure to antiretroviral nucleoside analogues. Lancet 1999;354:1084-1089. Guidelines for the use of antiretroviral agents in pediatric HIV infection. http://aidsinfo.nih.gov/ContentFiles/ElVIH DuranteElEmbarazoElPartoYDespuesDelParto_FS_sp.pdf; 2005. Tuomala RE, Shapiro DE, Mofenson LM, Bryson Y, Culnane M, Hughes MD et al. Antiretroviral therapy during

1129

A.P. Nso et al

26.

27. 28.

29. 30. 31.

32.

1130

pregnancy and the risk of an adverse outcome. N Engl J Med 2002;346:1863-1870. Culnane M, Fowler M, Lee SS, McSherry G, Brady M, O’Donnell K et al. Lack of long-term effects of in utero exposure to zidovudine among uninfected children born to HIV-infected women. Pediatric AIDS Clinical Trials Group Protocol 219/076 Teams. JAMA 1999;281:151-157. Bulletin E. Dysmorphology and Epidemiology Journal. http://bvs.isciii.es/mono/pdf/CIAC_03.pdf; 2004. Bermejo E, Martinez-Frias ML. Congenital eye malformations: clinical-epidemiological analysis of 1,124,654 consecutive births in Spain. Am J Med Genet 1998;75: 497-504. Berhman R, Kliegeman R, Jenson H. Nelson Pediatric Book. 17th ed. Madrid: Elsevier; 2004. Cruz M, Bosch J. Pediatric syndromes atlas. Barcelona: Espaxs; 1998. Cabrera N, Miranda T, Marrero L, Remón X, Cruz E. Congenital malformations of the hand: exploratory analysis of date. Rev Cubana Ortop Traumatol 1998;12:47-51. May M, Moreira A, García C et al. Pyelic ectasia of prenatal diagnosis. Incidence and study protocol during 2003. An

Pediatr (Barc) 2004;61:499-501. 33. Paredes Esteban RM, Salas Molina J, Ocana Losa JM, Garcia Ruiz M. [Immunohistochemical study of hypertrophic pyloric stenosis]. Cir Pediatr 2003;16:61-65. 34. Galbe J, group PP. Psychomotor development supervision and detection of autism disorders. http://www.aepap.org/ previnfad/pdfs/previnfad_desarrolloPM.pdf. 35. Campos-Castello J, de Santos Moreno MT, Jimenez-Alamo ML, Ramirez R, Careaga-Maldonado J. [Cerebral hemorrhage in full term newborns]. Rev Neurol 1999;29:239-246. 36. García N, Arias M. Nystagmus in children. www.esteve.es/ EsteveArchivos/1_8/Ar_1_8_51_APR_2.pdf. 37. Universal early detection of hypoacusis in Castilla la Mancha Program. www.jccm.es/sanidad/salud/promosalud/pdf/ PROGRAMA_HIPOACUSIA.pdf. 38. Massart F, Seppia P, Pardi D, Lucchesi S, Meossi C, Gagliardi L et al. High incidence of central precocious puberty in a bounded geographic area of northwest Tuscany: an estrogen disrupter epidemic? Gynecol Endocrinol 2005;20:92-98. 39. Marinkovic S, Bukarica S. [Umbilical hernia in children]. Med Pregl 2003;56:291-294.
