TISSUE DOPPLER IMAGING IN THALASSEMIA MAJOR PATIENTS: CORRELATION BETWEEN SYSTOLIC AND DIASTOLIC FUNCTION WITH SERUM FERRITIN

LEVEL Syarif Rohimi1, Najib Advani1, Sudigdo Sastroasmoro1, Bambang Madiyono1, Sukman Tulus Putra1, Mulyadi M Djer1, Fajar Subroto2

Abstract Background Thalassemia is a major public health problem in Indonesia. Cardiac diseases remain as the main cause of death in these patients due to iron overload. Although the T2* magnetic resonance; imaging has been considered as the gold standard for assessing cardiac iron overload but it has limited availability. The tissue doppler imaging (TDI) echocardiography, a fairly new and easy method that is suggested, can detect early abnormal myocardial iron overload. Objective To assess myocardial systolic and diastolic function of thalassemic patients using TDI and examine their correlation with serum ferritin level. Methods A cross-sectional study was conducted from January to March 2011 at the Harapan Kita Women and Children Hospital. We reformed clinical examination, serum ferritin level, as well as conventional and tissue doppler echocardiography on all subjects. Results We included 34 regularly-tranfused patients, -of which 17 were boys. The mean age of the subjects was 11.6 (SD 4.7 years, range 2.6 - 20 years). Mean pulse rate and blood pressure were within normal range. Hemoglobin level at inclusion ranged from 5.8 to 6 g/dL. Almost all patients did not receive regular chelation therapy. Median serum ferritin level was 6275 ng/mL (range 2151 17,646 ng/mL). Conventional echocardiography showed normal systolic function, but some diastolic dysfunctions were found including E wave abnormalites in 4 patients, A wave abnormalites in 3, and E/A ratio abnormalites found in 3. The TDI showed decreased systolic function (Sa wave abnormality) in 9 patients and diastolic dysfunctions (Ea wave abnormality in 11 patients and Aa wave abnormaly in 2). No abnormality was found in Ea/Aa and E/Ea ratios. There was a weak negative correlation between ferritin level and Sa wave and Ea wave

respectively and a moderately negative correlation between ferritin level and Ea/ Aa ratio. There was no correlation between serum ferritin and Aa wave or E/La ratio. Conclusion TDI identifies a greater number of patients with systolic and diastolic myocardial dysfunction than was revealed by conventional echocardiography. There was a weak negative correlation between serum ferritin to Sa wave and Ea wave, and a moderately negative correlation between ferritin and Ea/Aa ratio. There was no correlation between scrum ferritin and Aa wave or E/Ea ratio. [Paediatr Indones. 2012;52:187-93]. Keywords: tissue doppler imaging, echocardiography,-systolic function, diastolic function, ferritin, beta-thalassemia major Beta thalassemias are group of hereditary blood disorders characterized by anomalies in the synthesis of the beta chains of hemoglobin resulting in various phenotypes ranging from severe anemia to clinically asymptomatic individuals. In Indonesia, thalassemia is one of the most common single gene disorders and causes a major public health problem. Regular transfusion therapy and iron chelation are reported to improve patients' quality of life, but their life expectancy is limited by congestive heart failure associated with iron overload. Cardiac disease remains the main cause of death in patients with iron overload. Detection of early cardiac abnormality is difficult. The diagnosis of myocardial iron overload is often delayed because cardiac iron deposition is unpredictable. Also, symptoms and echocardiographic abnormalities arise late in the course of the disease. Usually, patients have normal exercise capacity, with systolic dysfunction occurring in the final state of disease. Tissue doppler imaging (TDI) is a fairly new and an easy method to detect abnormal myocardial iron overload in pediatric and adult patients with ^thalassemia. TDI has 88% sensitivity and 65% specificity compared to the gold standard. MRI T2*. The MRI is not widely available, time-consuming, and expensive, limiting its application in developing countries where thalassaemia is most common. In clinical practice, serum ferritin has been used to assess treatment effectiveness and is commonly used to assess the severity of iron

overload. The aim of this study was to assess cardiac systolic and diastolic functions using TDI and examine their correlation to serum ferritin in thalassemia patients with iron overload. Methods We conducted a cross-sectional study from January to March 2011 at the Harapan Kita Women and Children Hospital, Jakarta on children with beta thalassemia. Subjects with > 10 unit RBC tranfusion and agreed to participate in this study were enrolled and patients with congenital heart disease was excluded. All subjects underwent clinical examination, serum ferritin measurements, as well as conventional echocardiography and TDI according to the standardized methods. Outcome assessed were cardiac dysfunction. On conventional echocardiography, we measured left ventricular systolic function reflected by fractional shortening (FS) and ejection fraction (EF) as well as left ventricular diastolic function (the E wave, A wave, E/A ratio). The E wave reflected rapid ventricular filling phase while the A wave was the atrial contraction. The TDI measures myocardial systolic function (Sa wave), early myocardial diastolic function (Ea wave), late myocardial distolic function (Aa wave), Ea/Aa ratio, and E/Ea wave. TDI results was defined as abnormal if the Sa wave was <63 cm/ sec, Ea wave < 13 cm/sec, Aa wave <3,8 cm/sec, Ea/ Aa ratio < 2 and E/Ea ratio > 10. Sample size was calculated using the correlation coefficient formula using 5% level of significance.18 Numerical data was described as mean and standard deviation or median (range) as appropriate. Pearson's correlation analysis and simple linear regression were performed to assess the association between echocardiography /TDI parameters and ferritin level. Data was analyzed using SPSS version 11.5. Results We examined 34 regularly-transfused patients, of which 17 were boys. Baseline characteristics are shown in Table 1. Mean pulse rate and blood pressures were within normal limits. Conventional echocardiography showed normal systolic function, but some diastolic dysfunctions were found including E wave abnormalites in 4

patients, A wave abnormalites in 3, and E/A ratio abnormalites found in 3. The TDI showed decreased systolic function (Sa wave abnormality) in 9 patients and diastolic dysfunctions (Ea wave abnormality in 11 patients and Aa wave abnormaly in 2), as are shown in Table 2. Table 3 shows the distribution of TDI results. Pearson's correlation test revealed a weak correlation between serum ferritin level and Sa wave (r = 0.360, P = 0.036,). There was also a moderate correlation between serum ferritin level and Ea (r = 0.434, P = 0.010). No correlation was observed between serum ferritin level and Aa wave (r = -0.255, P = 0.146,), nor between serum ferritin level and E/Ea ratio (r=-0.174, P = 0.349). There was a moderately positive correlation between serum ferritin level and Ea/Aa ratio ( r=0.556, P=0.001). Linear regression analysis between serum ferritin level and Sa wave revealed an association (P= 0.036, r=0.36), with Sa increased 1 cm/sec for every ferritin level increase of 42048 cm/sec (Y=-18.12 + 420.48 X + e). An Sa wave of 13% was influenced by increasing serum ferritin level (Figure 1). Table 1. Subjects' characteristics Characteristics Age [years, median (range)] Age at diagnosis [months, median (range)] Boys (n=17) 10.1 (2.8-18.1) 4(2-72) Girls (n=17) 13(5.3-20) 8 (3-54) 8 (3-54) 34(15-43) 127(100-152) 0.87 (0.68-1.35) 6,1 (5.8-6.1) 101 (2) 105(5) 65(5) 9822(219917,646) 0 17

Age at first transfusion [months, median 4 (2-72) (range)] Body weight [kg, median (range)] Body height [cm, median (range)] Body surface area [m2, median (range)] Hemoglobin level [g/dL, median (range)] Heart rate [mean (SD)] Systolic pressure [mmHg, mean (SD)] Diastolic pressure [mmHg, mean (SD)] Serum ferritin [ng/mL , median (range)] Regular DFO administration, n Irregular DFO administration, n 24(14-45) 138(92-168) 0.85(0.58-1.53) 5.9 (6-6.2) 103(3) 100(7) 70(7) 6275(215111,000) 0 17

DFO: desferoxamine Table 2. Distribution of abnormalities based on echocardiography and TDI findings Echocardiography and TDI Left ventricle systolic function FS EF Left ventricle diastolic function E A E/A ratio Systolic myocardial function Diastolic myocardial function Sa wave Ea wave Aa wave 34 34 30 31 31 25 23 32 Normal n % 100 100 88 92 85 76 68 100 100 100 0 0 4 3 3 9 11 2 0 0 Abnormal n 0 0 12 8 8 26 32 6 0 0 %

Ea/Aa ratio 34 E/Ea ratio 34

Notes: FS = fractional shortening; EF = ejection fraction Table 3. Distribution of TDI results and serum ferritin levels
Ferritin levels (ng/mL) TDI parameter <2500 Sa wave E wave Aa wave E/Ea ratio Ea/Aa ratio 1 1 0 0 0 4 5 1 0 0 2501-5000 1 2 1 0 0 5001-7500 3 3 0 0 0 7501-10000 0 0 0 0 0 >10000 n 9 11 2 0 0 Number of subjects % 26 32 6 0 0

Linear regression analysis between serum ferritin level and Ea wave revealed a correlation (P= 0.010, r=0.434), Ea increased 1 cm/sec for every ferritin level increase of 429.965 ng/mL (Y = -47.98 + 429.97 X + e). An Ea wave of 13% was influenced by increasing serum ferritin level (Figure 2). In addition, linear regression analysis between serum ferritin level and Ea/Aa wave revealed a correlation (P= 0.001, r=0.556), with Ea/Aa ratio

increased 1 unit for every ferritin level increase of 13,517 ng/mL (Y= -25,090 + 13,517 X + e). A 30% decrease of Ea/Aa ratio was influenced by serum ferritin level (Figure 3). Figure 1. Correlation between serum ferritin and Sa wave Ferritin

Figure 2. Correlation between serum ferritin and Ea wave

Figure 3. Correlation between serum ferritin and Ea/Aa ratio Discussion

Discussion There is no passive excretory mechanism of iron. In thalassemia patients, iron is easily accumulated by repeated blood transfusions. Free iron, nontransferrin-bound iron (NTBI), and labile plasma iron in the circulation, as well as the labile iron pools within cells, are responsible for iron toxicity. Labile unbound iron is able to redox cycle between Fe2+ and Fe3 + , thereby generating reactive oxygen species (ROS), leading to lipid peroxidation and organelle damage. Ultimately this damage causes degeneration, fibrosis, cell death and dysfunction. In hemochromatosis cases, microscopic examination revealed large amounts of iron in muscle cells and histiocytes. Focal degeneration and fibrosis were extensive. Myocardial fibers varied in size and in iron content. In some, as much as two-thirds of the cell appeared filled with iron. The hearts were dilated as well as hypertrophied, in some to more than twice the expected weight. Quantifying myocardial iron content is difficult. The diagnosis of myocardial iron overload is often delayed because cardiac iron deposition is unpredictable. Symptoms and echocardiographic abnormalities arise late in the course of the disease. In clinical practice, serum ferritin has been commonly used to assess the severity of iron overload and the effectiveness of treatment. Serum ferritin does not correlate with myocardial iron load. Advantages of serum ferritin measurement are that it is easy to assess, inexpensive, amenable to serial measurements for monitoring chelation therapy, and correlates positively to

morbidity and mortality. However, serum ferritin measurements are not always reliable, since ferritin is an acute phase reactant and serum levels may be influenced by factors such as inflammatory disorders, liver disease and malignancy. Echocardiography studies have shown cardiovascular prognosis in p thalassemia patients to be excellent if serum ferritin is < 2500 ng/mL. Serum ferritin < 2500 ng/mL has been considered to be a safe level. A limitation in this study is that echocardiography and TDI were performed by only one person, possibly leading to a selective bias. However, this study should be viewed as preliminary study of TDI's correlation to the degree of it on overload in pediatric patients with P thalassemia. In our study, subjects's characteristics were similar to those in other studies (Table l).19'21 Indonesian subjects' characteristics were common to those found in subjects from other developing countries. The few differences in characteristics may have been due to the number of subjects, age distribution, compliance to chelation therapy or serum ferritin level. In our study, conventional echocatdiography showed no abnormalities in LV systolic function with normal fraction shottening (FS) and ejection fraction (EF). Kremastinos found that P-thalassemia was not a pure iron storage disease and the pathophysiology of cardiac dysfunction was poorly understood and multifactorial in etiology. Systolic dysfunction was. not correlated with serum ferritin and occured in a" late stage of the disease. Chronic iron myocardial deposition does not affect left ventricular relaxation. Engle et al. first reported in 1964 the co-existence of pericarditis and fatal arrhythmias with heart failure in pthalassemia major patients. Pericarditis usually coincided to some degree with myocarditis, as both are inflammatory heart diseases, usually with an immunological background. Patients with transfusion-dependent P-thalassemia are at risk of acquiring viral infections such as hepatitis B and C, as well as IIIV. The increased frequency of infections associated with p-thalassemia seems to be related to abnormalities of the immune system. Multiple transfusions represent a repetitive antigenic stimulus, together with iron chelation therapy itself. Once

systolic function of the left ventricle becomes impaired, survival is reduced, suggesting it to occur at a very late stage in the disease process. Diastolic LV dysfunction measurements included E wave > 0.70 m/sec in 4 subjects, A wave < 0.30 m/sec in 3 subjects and E/A ratio < 2 in 3 subjects (Table 2), findings which were similar to those of other studies. Diastolic dysfunction proceeded systolic dysfunction. Early restrictive diastolic function occurred and correlated to severe serum ferritin level. TDI is a relatively new and an easy method with 88% sensitivity and 65% specificity compared to MRI T2* for earlier detection of abnormal myocardial iron overload.10 TDI is superior and reproduceable in detecting myocardial dysfunction. Silvilairat et al. evaluated TDI in 31 patients with normal LVFS and found that diastolic myocardial dysfunction was absent in all patients with serum ferritin < 2500 ng/ mL, but was present in all patients with setum ferritin > 5000 ng/mL.11 We found that TDI revealed systolic myocardial dysfunction (Sa wave) in 26% of those subjects in which conventional echocardiography did not show any abnormalities. TDI also showed decreasing left ventricular myocardial diastolic function (Ea wave) in 32% of subjects and decreasing atrial contraction (Aa wave) in 6% of subjects. No abnormalities were found in Ea/Aa and E/Ea ratios of all subjects. There was a weak negative correlation between ferritin and Sa wave and Ea wave, but a moderately negative correlation between ferritin and Ea/Aa ratio (Figures 1-3). There was no correlation between serum ferritin and Aa wave or E/Ea ratio. Using TDI, we observed that diastolic dysfunction was more pronounced than systolic dysfunction. Iron deposition in the heart may be patchy and not uniform. Iron is known to accumulate in the ventticular septum, as well as the free wall of the ventricles, with a tendency to be more concentrated in the epicardial layers.27 Clinical and experimental studies have shown that iron is deposited within myocytes rather than within the interstitium. Wall motion abnormalities may represent an early sign of cardiac disease despite preserved global function. The regional abnormalities are related to iron overload and are easily detectable with TDI. The regional wall motion in patients with thalassaemia and iron

overload is altered in the absence of global dysfunction. It is possible that at an early stage iron is predominantly deposited in the septum while at a later stage other areas become affected. Long-term prospective studies of echocardiographic assessment for systolic and diastolic ventricular function in larger numbers of pediatric patients with P-thalassemia is warranted. TDI should be a routine part oi the echocardiographic assessment of pediatric patients with p-thalassemia. In conclusion, TDI revealed a greater number of patients with systolic and diastolic myocardial dysfunction than was revealed by conventional echocardiography. There was a weak negative correlation between serum ferritin and Sa wave and Ea wave, and a moderately negative correlation between ferritin and Ea/Aa ratio. There was no correlation between serum ferritin to Aa wave or E/Ea ratio. References 1. Galanello R, Otiga R. Beta-t'.ialassemia. Orphanet ] Rare Dis. 2010;5:11-36. 2. Olivieri NF. Medical progress: the B-thalassemias. N Engl J Med. 1999;341:99-109. 3. Rund D, Rachmilewitz E. Medical progree: ^-thalassemia. N Engl] Med. 2005;353:1135-46. 4. Indonesian Ministry of Health. Thalassemia masalah serius kesehatan masyarakat [serial online], [cited 2011 Jan 15]. Available from:

http://www.depkes.go.id 5. Engle MA, Erlando M, Smith CH. Late cardiac complies tions of chronic, severe, refractory anemia with hemochromatosis. Circulation. 1964;25:698705. 6. Olivieri NF, Nadian DG, MacMillan JH, Wayne AS.vLW PR McGee A, et al. Survival in medically treated patients with homozygous a-thalassemia. N Engl Med. 1994;331: 574-8. 7. Kremastinos DT, Tiniakos G. Theodorakis GN, Katritsis DG, Toutouzas PK. Myocarditis in B-thalassemia major. Circulation. 1995;91:66-71.

8. Leon MB, Borer JS, Bacharach SL, Green MV, Benz EJ, Griffith P, et al. Detection of early cardiac dysfunction in patients with severe beta-thalassemia and chronic iron overload. N Engl] Med. 1979;301:1143-8. 9. Cogliandro T, Derchih G, Mancusoc L, Mayer MQ Pannone B, Pepe A, et al. Guideline tecommendations for heart complications in thalassemia major. ] Cardiovasc Med. 2008;9:515-25. 10. Vogel M, Anderson LJ, Holden S, Deanfield JE, Pennell DJ, Walker JM . Tissue Doppler echocardiography in patients with thalassemia detects early myocardial dysfunction related to myocardial iron overload. European Heart Journal. 2003;24:113-9. 11. Silvilairat S, Sittiwangkul R, Pongprot Y, Chatoenkwan P Phornphutkul C. Tissue Doppler echocardiography reliably reflects severity of iron overload in pediatric patients with beta thalassemia. Eur J Echocardiogr. 2008;9:368-72. 12. Jabbar DA, Davison G, Muslin AJ. Getting the iron out: preventing and treating heart failure in transfusion-dependent thalassemia. Cleve Clin] Med. 2007;74:807-16. 13. Garadah TS, Kassab S, Mahdi N, Abu-Taleb A, Jamsheer A. Pulsed and tissue Doppler echocardiographic changes in patients with thalassemia major. Clin Med Insights: Blood disorders. 2010;3:1-8. 14. Snider AR, Sewer EA, Rutter SB. Echocatdiography in pediatric heart disease. St Louis: Mosby; 1997. p. 195-226. 15. Hills JC, Palma RA. Doppler tissue imaging for the assessment of left ventricular diastolic function: a systematic approach for the sonographer. J Am Soc Echocardiogr. 2005;lG:80-8. 16. Ho CY, Solomon SD. A clinician's guide to tissue Doppler imaging. Circulation. 2006;113:396-8. 17. Kapusta L, Thijssen JM, Cuypers MHM, Peer PGM, Daniels O. Assesment of myocardial velocities in healthy children using tissue Doppler imaging. Ultrasound Med Biol. 2000;26:229-37.

18. Madiyono B, Moeslichan MZ, Sastroasmoro S, Budiuian HI, Purwanto SH. Perkiraan besar sapel. In: Sastroasmoro S, Ismael S, editors. Dasar-dasar metodologi penelitian klinik. 3rd ed. Jakarta: Sagung Seto; 2008. p. 302-30. 19. Subroto F, Munthe BG, Advani N, Firmansyah A. The correlation between ferritin level and cardiac dysfunction in patients with thalassemia. Pacdiatr Indonesia. 2003;43:24-7. 20. Suwrniaty R, Ontoseno T, Permono B, Sudigdo S. Pengaruh kadar feritin serum terhadap fungsi ventrikel kiri pada thalassemia B mayor yang mendapat transfusi multipel. Sari Pediatri. 2007;9:178-84. 21. Ali M, Putra ST, Gatot D, Sastoasmoro S. Left ventivular functions and mass of the adolescents and yourg adults with thalassemia major: An echocardiography study. Paediatr Indones. 2006;46:214-9. 22. Lau KC, Li AMC, Hui PW, Yeung CY. Left ventricular function in p thalassaemia major Arch Dis Child. 1989;64:1046-51. 23. Spirito P, Lupi G, Melevendi C, Vecchio C. Restrictive diastolic abnormalities identified by Doppler echocardiography in patients with thalassemia major Circulation. 1990;82:88-94. 24. Ktemastinos DT, Tsiapras DP, Tsetsos GA, Rentoukas EI, Vretou HP Toutouzas PK. Left ventricular diastolic Doppler characteristics in athalassemia major. Circulation. 1993;88:1127-35. 25. Atiq M, Bana M, Ahmed US, Bano S, Yousuf M, Fadoo Z, et al. Cardiac disease in beta-thalassaemia major: is it reversible? Singapore Med J. 2006;4:693-6. 26. Khan FR, Mahsud MAJ, Ayub T, Khan MH, Shah SH. Frequency of heart failure in patients with B thalassemia major. Gomal J Med Sci. 2006;2:49-51. 27. Arshad MS, Hyder SN. Evidence of abnormal left ventricular function in patients with thalassemia major: an echocardiography based study. ] Ayub Med Coll Abbottadebad. 2009;21:37-41. 28. Papadopoulou-Legbelou K, Varlamis SG, Metaxa MA, Karamperis S, Zafiriou AM. Full resting echocardiographic study of left ventricle in children with b-thalassemia major. Cardiologia. 2009;2:132-8.

29. Walker JM. The heart in thalassemia. Eur Heart ]. 2002:2;102-5. 30. Kohgo Y, Ikuta K, Ohtake T, Torimoto Y, Kato J. Body iron metabolism and pathophysiology of iron overload. Int J Hematol. 2008;88:7-15. 31. John Porter J. Iron toxicity and clinical sequelae [serial online], [cited 2011 Jan 15]. Available from: http://www. pdf-searcher.com/Iron-Toxicity-andClinical-Sequelae-John-Porter,-MA,-MD,-FRCEhtml 32. Mohan JC, Reddy KS. Thalassemic cardiomyopathy: current concepts. Indian J Pediatr. 1984;51:383-6. 33. Kremastinos DT. Iron overload and myocardial restriction. Hospital Chronicles. 2007;2:140-2. 34. Inati A, Zeineh N, Ismaeel H, Koussa S, Gharzuddine W, Taher A. (3thalassemia: the Lebanese experience. Clin Lab Haem. 2006;28:217-27.

EFFICACY OF SALBUTAMOL-IPRATROPIUM BROMIDE NEBULIZATION COMPARED TO SALBUTAMOL ALONE IN CHILDREN WITH MILD TO MODERATE ASTHMA ATTACKS Matahari Harumdini, Bambang Supriyatno, Rini Sekartini

Abstract Background The efficacy of salbutamol-ipratropium bromide nebulization in children with moderate asthma attacks remains unclear, and studies on patients with mild attacks have been relatively few, especially in Indonesia. However, it is common practice for this drug combination to be given to patients with mildmoderate asthma attacks. Objective To compare the efficacy of salbutamol-ipratropium bromide nebulization to salbutamol alone in children with mild to moderate asthma attacks. Methods This single-blind, randomized clinical trial was held in the Department of Child Health at Cipto Mangunkusumo Hospital, the Tebet Community Health Center, and the MH Thamrin Salemba Hospital on children aged 5-18 years with mild to moderate asthma attack. We randomized subjects to receive either 2.5 mg salbutamol plus 0.5 mg ipratropium bromide (experimental group) or 2.5 mg salbutamol alone (control group). Nebulization was given twice, with a 20 minute interval between treatments. We assessed clinical scores, vital signs, oxygen saturations, and peak flow rates (PFRs) at baseline, and every 20 minutes up to 120 minutes post-nebulization. Results A total of 46 patients were randomized to either the experimental or the control group. Subjects had similar baseline measurements. At 20 minutes postnebulization, the percentage increase of PFR was 19% higher in the experimental group (P=0.01,95% CI 1.8 to 47.2). The proportion of PFR reversibility was 27% higher in the experimental group, although this result was statistically insignificant (P=0.06, 95% CI 0.03 to 0.52). There were no significant differences in clinical scores, oxygen saturations, respiratory rates, or hospitalization rates between the two groups. Side effects also did not differ significantly. Conclusion Salbutamol-ipratropium bromide nebulization improved PFR measurements better

than salbutamol alone. However other clinical parameters were not significantly different between the two groups. [Paediatr Indones. 2012;52:200-8]. Keywords: children, mild to moderate asthma attack, ipratropium bromide, salbutamol

Asthma is global health problem in children, and is increasing in prevalence, even though the pathogenesis, pathophysiology, and management of asthma is well understood. The National Health Interview Survey in the United States reported an asthma prevalence of 7.5% in 1995.1 In Indonesia, Rahajoe et.al. reported asthma prevalence to be 6.7%.2 Controversies in asthma management may increase morbidity and mortality of patients. The addition of ipratropium bromide for patients with asthma attacks has been controversial. Beta2- agonists are potent bronchodilators, but multiple or large doses may cause adrenergic side effects. However, ipratropium bromide is an anticholinergic bronchodilator with a slower onset, longer duration of action, and less adrenergic side effects compared to those of beta2-agonists. Previous studies have shown that a combination of salbutamol and ipratropium in patients with severe asthma attacks improve lung function and clinical score, while lowering emergency department (ED) admission duration and hospital admission rates. Other studies have also reported salbutamol-ipratropium bromide superiority in patients with moderate attacks, while studies on its use in patients with mild asthma attacks have been few. Salbutamol-ipratropium nebulization has commonly been given to patients with mild to moderate asthma attacks, although only one Indonesian study to date has been published on this subject. We aimed to compare the efficacy of salbutamol-ipratropium nebulization with salbutamol alone in pediatric patients with mild to moderate asthma attacks. We measured and compared clinical scores, peak flow rates, oxygen saturations, respiratory rates, and hospital admission rates of the two groups. Methods

This study was designed as a single-blind, randomized, clinical trial performed from September 2010-March 2011 at the Community Health Center of Tebet District, and the EDs of Cipto Mangunkusomo Hospital and MH Thamrin Salemba Hospital. We compared the effects of nebulization with salbutamolipratropium combination to those of salbutamol alone. Patients aged 5-18 years who visited the ED with mild to moderate asthma attacks, classified according to Schuh's asthma clinical score, were eligible for enrollment. We excluded patients with signs of respiratory failure, need of intensive care management, heart abnormality, pneumonia or other respiratory disorders altering lung function, ocular disorder altering intraocular pressure or pupillar response (as diagnosed by history-taking and physical examination), hypersensitivity to ipratropium or salbutamol, and those who had received ipratropium bromide treatment within the 36 hour;: prior to enrollment. Subjects' parents provided informed consent. We consecutively assigned subjects to receive either salbutamolipratropium bromide (experimental group) or salbutamol alone (control group), according to a drug sequence table generated by block randomizations of six. This table was kept by the principal investigator (PI) to keep the subjects blinded to their allocated group. Subjects were given either 2.5 mg salbutamol with 0.5 mg ipratropium bromide (Combivent) or 2.5 mg saibutamol (Ventolin) nebulization in 3-5 ml saline. Subjects were given two doses by ultrasonic nebulizer (Omron NE-C29) via face mask, with a 20 minute interval between treatments. The duration of each nebulizer treatment was about 10 minutes. At enrollment, subjects' baseline data was collected including demographic characteristics (age, sex, and nutritional status), asthma history, treatment history, asthma comorbidities (allergic rhinitis or sinusitis), duration of current symptoms, and asthma severity. We also measured baseline clinical parameters, including Schuh's clinical scores, vital signs, PFRs by mini peak flow meter (Breath-Taker, Australia, reproducibility 8.4%, SD 27 L/m), and oxygen saturation by pulse oxymetry (Oxy3, OneMed). Clinical response was assessed every 20 minutes, until 2 hours post-nebulization,

including the same parameters measured at baseline. For patients with moderate attacks, we planned to also measure blood gas analysis (BGA) twice, at baseline and at 2 hours after treatment, though most patients declined. PFR was measured by forced expiration maneuver (patient twice performed forced expiration after maximal inspiration with at least a 5-second interval between forced expiration). Only the best value was recorded. Patients with inadequate clinical response after 2 hours post-treatment were admitted to the hospital. If the principal investigator (PI) was absent when an asthma attack patient came to ED, the clinical score at baseline was measured by a research assistant or by trained ED attending physicians. When a subject enrolled, the PI was called by phone for study group random allocation instructions. By the time the second nebulization was finished, the PI would have arrived at the ED to continue data measurements. Prior to the study, interrater reliability for baseline clinical scoring WHS assessed by comparing the Pi's ratings to those of a research assistant on 6 patients. Individual severity scores were summed and divided into three severity groups as follows: mild (total score 1-3), moderate (total score 4-6) or severe (total score 7-9) (Table 1). Interrater reliability was measured using these severity subgroups, with Kappa = 0.6. The primary outcome was nebulization efficacy, measured by several parameters including decreased clinical score, increased PFR, increased oxygen saturation, decreased respiratory rate, and decreased percentage of hospital admission. PFR was measured as the percentage increase from baseline. PFR reversibility was defined as a PFR increase >12% from baseline. The proportion of patients with PFR reversibility in each group was also recorded. Secondary outcomes were blood gas values before and after treatment, and side effects of medications. The required sample size was determined by a formula of mean difference of two independent groups, with oc=5% and power of 80%. Since a previous study showed that the standard deviation of mean change in clinical asthma score between the saibutamol-ipratropium and salbutamol groups was 1.5,9 the

clinically significant difference was set at 1.5. Therefore, 16 patients per study group, or a total of 32 patients, were needed for this study. Differences in clinical scores, PFRs, oxygen saturation, and respiratory rates between groups were analyzed by independent t-test, or Mann-Whitney test if the data had an abnormal distribution. Differences in PFR reversibility, hospital admission, and side effects were analyzed by Chi-square test or Fisher's exact test. We performed intention-to-treat analyses and considered P<0.05 to be statistically significant. This study was approved by The Medical Research Ethics Committee, Faculty of Medicine, University of Indonesia. Results A total of 46 patients were enrolled after 6 patients were excluded due to severe asthma attack (3 patients), had received ipratropium bromide within the prior 36 hours (2 patients) or had a heart abnormality (1 patient). Of the 46 subjects, 32 had mild asthma attacks (16 patients were allocated to each group), while 14 patients had moderate asthma attacks (7 were allocated to each group). All baseline parameters were similar between the two groups (Table 2). At 40, 60, and 120 minutes after nebulization, clinical scores decreased more in the experimental group, but they were not statistically or clinically different from the control group by Mann-Whitney test (Table 3). In subjects with moderate attacks (n= 14), we found an apparent difference in the mean decrease of clinical scores between the two groups. The 1.58 point difference between the two groups seemed substantial, but statistical significance could not be established because of inadequate sample size (Table 3). PFR data was analyzed for 40 patients, since 6 patients failed to complete PFR measurements due to their clinical conditions. At 20-120 minutes, we found higher PFR percentage increases from baseline in the experimental group than in the control group. The median difference at 20 minutes was 19% (95%CI 1.80 to 47.18; P=0.012) and the median differences at 40, 60, and 120 minutes were all 25% (data was abnormally distributed) (Table 4). Statistical significance was observed only at the 20-minute time point (P=0.012). In mild attack subjects alone

(n=30), we found mean difference of 15,8% at 20 minutes (95% CI 1,05 to 30,31; P=0,05; Table 4). In moderate subjects alone (n=10), we found more than 50% mean difference at the beginning and final observation, but statistical analysis could not be performed (Table 4). Table 1. Schuh's clinical asthma score 9
Score 0 Accessory muscle score No retractions Wheeze score No wheeze and moving air well 1 Intercostal retractions End-expiratory wheezes 2 Intercostal and suprasternal retractions 3 Nasal flaring Panexpiratory + inspiratory wheezes Wheezes audible without stethoscope Normal activity and speech; minimal dyspnea Decreased activity; 5-8 word sentences; moderate dyspnea Concentrate on breathing; <5 word sentences; severe dyspnea Dyspnea score Dyspnea absent

Mild attack: total score 1-3; moderate attack: total score 4-6; severe attack: total score 7-9

Table 2. Subjects' baseline demographics, asthma history and clinical parameters

Parameters Control (n=23) Experimental (n=23) Mean age, years (SD) Male gender, n Obese, n Onset of asthma > 5 years ago, n Asthma severity: infrequent attack episodes, n Asthma attack severity: mild attack, n Median duration of current symptoms, days (range) Previous prophylaxis, n Current non-steroid, non- ipratropium inhalant use, n Current steroid use, n Asthma comorbidities (allergic rhinitis and/or sinusitis), n Median initial clinical score, (range) Median initial peak flow rate, liters/min (range) Median initial oxygen saturation, % (range)

11.39(2.56) 12 1 13 14 16 2(1-7) 2 10 2 8 2(1-4) 175(50-350)

11.07(3.29) 11 2 13 15 16 1 (1-7) 3 7 2 5 3(1-6) 100(50-300)

Mean initial respiratory rate, x/min (SD) Mean initial heart rate, x/min (SD)

97 (96-99) 28.76 (6.24)

98 (96-99) 29.42(7.11)

102.19(12.54) 103.74(20.03)

Table 3. Comparative median/mean decreases in clinical score

Decrease in clinical score Subjects Time at evaluation Control (n= 23) median (range) Mild to moderate attack 20 minutes 40 minutes 60 minutes 120 minutes 2(1-4) 2(1-5) 2(1-5) 2 (1-5) Control (n=7) mean (SD) Moderate 20 minutes 2.14 (0.90) 3.57 (0.98) 3.71 (0.95) 3.71 (0.95) Experimental (n=23) median (range) 2(1-4) 3(1-6) 3(1-6) 3(1-6) Experimental (n= 7) mean (SD) 2.43 (0.98) 4.14 (1.07) 4.86 (1.07) 5.29 (0.76) Statistical analysis was not performed due to inadequate number of subjects 0.560 0.775 0.524 0.414 -0.40 to 0.57 -0.55 to 0.81 -0.42 to 1.12 -0.34 to 1.29 P 95% CI

attacks only 40 minutes 60 minutes 120 minutes

We observed a higher proportion of subjects with PFR reversibility in the experimental group (17/19 subjects) than in the control group (13/21 subjects) at 20 minutes. The difference between groups was 0.27 (95% CI0.026 to 0.524; P=0.069). At 40 to 120 minutes, there were similarly no significant differences in proportions of subjects with PFR reversibility (Table 5). In subjects with mild attacks alone (n= 30), there was also no difference between groups (Table 5), while in subjects with moderate attacks alone (n= 10), reversibility tended to be higher in the experimental group, but the sample size was too small to analyze (Table 5). Before intervention, all subjects had oxygen saturation >95%, therefore, oxygen therapy was not needed. There were no differences in oxygen saturation between the two groups. We also found no significant difference in the decrease of respiratory rates between the two groups (Table 6) In moderate asthma attack

subjects alone (n= 14), there was a greater decrease in respiratory rates in the experimental group. These differences were 4 x/minutes at 20 and 40 minutes, and 6x/minutes at 60 and 120 minutes, which were clinically quite apparent (Table 6) but statistical analyses were not performed due to lack of subjects. Two patients with moderate asthma attack from the control group responded inadequately at 120 minutes, requiring hospital admission. These 2 subjects were given ipratropium bromide nebulization and intravenous steroids. They were analyzed in the control group, since we used intention-to-treat in the experimental group (0/23 subjects), but this analyses. The number of hospital admissions was difference was not statistically significant (P = 0.489) higher in the control group (2/23 subjects) than (Table 7). Table 4. Comparative median FFR percentage increase
Subjects Time at evaluation 20 minutes 40 minutes 60 minutes 120 minutes Control (n=21) % increase, (range) 14.28(0-100) 25(7.14-100) 25(11.1-100) 25(11.1-100) Control (n=16) % increase, (range) Mild attacks only 20 minutes 40 minutes 60 minutes 120 minutes 13,39(0-57,14) 22,5(7,14-60) 25(11,11-100) 25(11,11-100) Control (n=5) % increase, (range) Moderate attacks 20 minutes only 40 minutes 60 minutes 120 minutes 16,7(0-100) 50(16,7-100) 50(16,7-100) 50(i6,7-100) Experimental (n=19) % increase, (range) 33.33 (8.33-200) 50 (8.33-200) 50 (8.33-300) 50 (8.33-200) Experimental (n=14) % increase, (range) 29,16(8,33-100) 50(8,33-100) 50 (8,33-200) 50 (8,33-200) Experimental (n=5) % increase, (range) 66,7 (20-200) 100 (20-200) 166,7(60-300) 166,7(60-300) Statistical analyses was not performed due to lack of subjects 0.012 0.114 0.115 0.115 P* 0.058 0.234 0.531 0.531 1.80 to 47.18 1.61 to 54.17 4.46 to 83.59 4.46 to 83.61 95% CI 1.05 to 30.31 -2.5 to 40.96 -10.56 to 49.46 -10.56 to 49.46 P* 95% CI

Mild to moderate attack

*Mann-Whitney test. CI was measured by formula using mean v&;us

Table 5. Comparative proportions of PFR reversibility (defined as PFR increase >12% from baseline)
Reversibility 5t Mild to moderate 20 minutes attacks 40 minutes 60 minutes 120 minutes Control (n=21) 13 19 20 20 Control . (n=16) Mild attacks only 20 minutes 40 minutes 60 minutes 120 minutes 10 14 15 15 Control (n=5) Moderate attacks 20 minutes only 40 minutes 60 minutes 120 minutes 5 5 5 5 Experimental (n= 19) 17 17 17 17 Experimental (n= 14) 12 12 12 12 Experimental (n=5) 5 5 5 5 Statistical analysis was not performed due to lack of subjects P* 0.069q 1q 0.596 0.596 P 0,226a 1
b q q

95% CI 0.026 to 0.524 -0.409 to 0.431 -0.107 to 0.223 -0.107 to 0.223

Not measured

0,586b 0,586b

a Chi-square test b Fisher's exact test

Table 6. Comparative median respiratory rate decrease

Time at Evaluation Control (n=23) Experimental (n= 23) P* 95% CI

Mild to moderate attacks

20 minutes 40 minutes 60 minutes 120 minutes

4 (2-18) 6 (4-20) 6 (4-20) 6 (4-20)

Control (n=7)

4 (0-8) 8 (1-16) 8 (1-16) 8 (1-16)

Experimental (n= 7)

0,907 0,585 0,602 0,602

-2,86; 1,12 -2,19; 2,79 -2,40; 3,19 -2,41; 3,18

Moderate attacks only

20 minutes 40 minutes 60 minutes

4 (4-16) 6 (4-16) 6 (4-18)

8 (4-8) 10 (8-12) 12 (8-12)

Statistical analysis was not performed due to lack of subjects

120 minutes

6 (4-18)

12 (8-12)

* Mann-Whitney test; CI measured by formula using mean value

Table 7. Comparative proportion cf hospitalization

Control (n=23) Experimental (n= 23) 0 23

Hospitalization No hospitalization P=0.488 ( Fisher's exact test)

2 21

Of 14 patients with moderate attacks, only 2 consented to BGA examination. The first patient agreed to arterial puncture after the intervention, and the second patient agreed before the intervention. In both subjects, we found decreased pressure of oxygen in arterial blood (Pa02) (33.6 and 31 mmHg, respectively) and low HC03 (21 and 19 mmol/L, respectively), while pH, Pa02 and oxygen saturation were still normal. We found the side effect of mouth mucosal dryness to be of similar proportions in both groups. The unilateral decrease of light pupillar response was found in 2 patients from the experimental group at 20 minutes, but spontaneously resolved at 40 minutes. The proportion of subjects with tachycardia was highest at 20 minutes, but did not differ between groups. Tachycardia resolved with time. Discussion This study had some limitations. In this single-blinded study, investigators were not blinded, but subjects were. Ideally, the study should be double-blinded, since we used a subjective parameter of efficacy (clinical score). However, the other efficacy parameters (PFR, oxygen saturation, respiratory rate, and proportion of hospital admission) were objectively measured. Also, the PFR could net be measured in 6 patients, but the remaining sample size was still adequate for most statistical analyses. In addition, we planned to measure BGA in all subjects with moderate attacks, but most subjects refused the arterial puncture. Since Carruthers et al. showed that respiratory failure was unlikely in patients with oxygen saturation >92%, BGA was not necessary unless otherwise clinically indicated.

In our study, interrater reliability could only be measured between the Pi and a research assistant on 6 patients, due to limitations of time and sample size. Our Kappa was 0.6 (0.6-0.8 was considered sufficient). The same clinical score was used by previous studies with Kappa values ranging from 0.6 to 0.9, similar to that of our study. The required minimal sample size was 32 subjects, but at the end of the study, we had more subjects to be analyzed. We attempted to subgroup analyses for different attack severities. However, a subanalysis could only be performed on subjects with mild attacks due to insufficient number of subjects with moderate attacks. Therefore, we analyzed data of mild and moderate attack subjects as a whole, while trying to demonstrate clinical differences in each subgroup. The low number of asthma attack patients at our public facilities may be due to increasing numbers of private health centers with nebulization facilities as well as better maintenance treatment for asthma patients.

Demographic and clinical parameters that may influence the clinical response to nebulization treatment were assessed at baseline, and found to be similar in the two groups. We observed an insignificant difference in clinical scored throughout che study between the two groups (median difference of 1 point). Similarly, Rayner et al. reported that ipratropium bromide given after beta2-agonist resulted in a reduced synergistic effect. Furthermore, Kumaratne et al.11 reported that in young subjects (4 months-6 years) assumed to have a predominant bronchospasm on peripheral small bronchi, ipratropium bromide was less effective. In our analysis of subjects with moderate attacks alone, we found a greater decrease in clinical score in the experimental group than in the control group (mean difference 1.58 points), though further statistical analyses could not be performed due to insufficient subjects. Previous studies by Schuh et al. Sharma et at. Kartininingsih et al. and Qureshi et al. also demonstrated a larger decrease in clinical score in their experimental groups. Those studies included children of younger age and greater numbers of subjects with moderately severe attacks. Our

study included mostly subjects with mild asthma attacks, in which less cholinergic activity occurs. On the other hand, in subjects with moderate attacks, we found a larger difference in the decrease in clinical scores. This difference might have been more profound if the number of subjects with moderate attacks was larger. The Global Initiative for Asthma (GINA) recommends lung function tests to confirm diagnoses and to evaluate asthma severity, as well as asthma attack severity. Lung function tests generally comprise spirometry and peak flow meter examinations. We chose to use peak flow meters due to their greater availability. The reference data for predicting PFR values for age, sex, and body mass index in patients aged 5-18 years in Indonesia was insufficient, so we gauged PFR response to be the percentage of increase from baseline, and the proportion of patients with PFR increase of J> 12% from baseline (PFR reversibility). We found a 19% difference (95%CI 1.80to47.18; P=0.012) in PFR between the groups at 20 minutes. Beyond 20 minutes, we also found differences of 25%, but they were not significant. However, the increasing confidence interval suggested relevant differences beyond 20 minutes. This result was consistent with previous studies9'20 which showed more profound differences of lung function parameters at the end of the observations, due to the slower onset of ipratropium bromide compared to that of salbutamol. Similarly, in a meta-analysis on subjects with moderate to severe attacks, Rodrigo et al. found a difference of 12.4% in forced expiratory volume at 1 minute (FEVj) measured by spirometry. Sharma et al. also found a higher PFR increase percentage in the experimental group at 30 minutes to 4 hours after nebulization, in a study on subjects with moderate attacks. In subjects with moderate attacks alone, we found a larger difference in PFR improvement (>50%) at 20 to 120 minutes, but statistical analyses could not be performed. Schuh et al. reported that differences in FEV increased as attack severity increased. Nonetheless, a significant difference in PFR increase was reported by Rayner et al. who gave ipratropium bromide sequentially after salbutamol, and Qureshi et al.8 who completed PFR measurements in only 40% of their subjects.

We also found a greater proportion of PFR reversibility at 20 minutes in the experimental group. The difference in proportion was only 27% (95%CI 0.026 to 0.524; P=0.069), in contrast to previous studies which showed better efficacy at the end of the observation periods. Our study included mostly patients with mild attacks and less bronchoconstriction, thus the synergistic effect of ipratropiumsalbutamol was observed at just 20 minutes. At the subsequent rime points, the proportion of reversibility did not further increase because maximal

bronchodilatation had already occurred at 20 minutes. Kartininingsih et al. and Qureshi et al. found significant differences in oxygen saturation between their groups, in subjects with moderate to severe attacks. In contrast, most of our subjects had mild attacks with high oxygen saturation (96-99%) at baseline, thus clinical improvement could not be shown. Ducharme et al. also reported no significant difference in oxygen saturation in subjects with mild to moderate attacks. We observed no significant difference in decreased respiratory rates between the two groups. However, in moderate attack subjects alone, we only found a tendency of difference between the groups. Studies by Sharma et al. and Qureshi et al. reported a greater decrease in respiratory rate in the experimental groups, in subjects with moderate attacks and subjects with severe attacks, respectively. Our contrasting results may be due to the smaller number of subjects with moderate attacks in our study. Many previous studies on subjects with moderate to severe attacks reported lower hospital admission rates in the ipratropium bromide group. The difference in admission rates between groups was greatest in the most severe cases. We found a small difference in hospital admissions (2/23 in the control group vs 0/23 in the experimental group), but it was not statistically significant (P=0.489). Most of our subjects had mild attacks, and as such were less likely to be hospitalized. Furthermore, our sample size was too small to detect any differences in hospitalization rates. An asthma attack patient may initially hyperventilate to increase oxygen uptake, thus decreasing carbon dioxide levels. If the obstruction continues, the

ventilation-perfusion mismatch can no longer be overcome by hyperventilation, thus resulting in hypoxemia and hypercapnia. Carruthers et al. reported that the respiratory failure rate was only 4-2% among patients with oxygen saturation > 92%. In contrast, in patients with oxygen saturation <92%, 29,4% had respiratory failure. In our study, both subjects that we performed BGA on had hypocapnia and normal oxygen saturation, consistent with previous studies. The relatively low value of HCC3 revealed a tendency towards metabolic acidosis which can be caused by the increase of plasma lactate due to increased respiratory muscle activity under hypoxic conditions. Beta2-agonist receptor stimulation may also generate gluconeogenesis, glycolyis and lipolysis, producing lactate. The two subjects who had BGA assessed in this study did not show clinical signs of metabolic acidosis; despite the low value of HCO3, thus they did not need any additional specific management. There were 2 patients with pupil abnormalities in the experimental group. These side effects were reversible. Mouth mucosal dryness did not differ between the groups. Tachycardia was also similar between the groups, and resolved with time. Tachycardia was not only due to side effects of medications, but was also a physiologic response to mismatched ventilation-perfusion, resolving asclinical condition improved. Qureshi et al. Ducharme et.al. and Rodrigo et al. also reported no significant differences in side effects with the addition of ipratropium bromide.Despite the study limitations, we conclude that salbutamol-ipratopium bromide nebulization showed better efficacy compared to salbutamoi alone in patients with mild to moderate asthma attacks. The PFR percentage increase and PFR reversibility at 20 minutes was better clinically for the experimental group. However, other clinical parameters of efficacy (clinical scores, oxygen saturation, respiratory rates, and hospital admission rates) were not different between groups. In subjects with moderate attacks alone, we observed a tendency to better efficacy with the addition of ipratropium bromide, based on clinical score, PFR, and respiratory rate. Nevertheless, further studies with a larger sample size for subjects with moderate attacks are necessary. Acknowledgment

We thank Boebringer Ingelheim for supporting the study, as well as EM Dadi Suvoko, MD, Partini R Trihono, MD, Antonius H. Pudjiadi, MD, and Waldi Nurhamzah, MD, for their valuable advices. We also thank our research assistant, Hari Nugroho, MD and all staff of the Respirology Division of the Department of Child Health at Cipto Mangunkusumo Hospital, the Cipto Mangunkusumo Hospital ED, the MH Thamrin Salemba Hospital ED, and the Community Health Center of Tebet District. References 1. Akinbami LJ, Schoendorf KC. Trends in childhood asthma: prevalence, health care utilization, and mortality. Pediatrics. 2002;110:315-22. 2. Rahajoe N, Supriyatno B, Setyanto DB. Pedoman nasional asma anak. Jakarta: UKK Respirologi PP Ikatan Dokter Anak Indonesia; 2004. p. 3-4. 3. Lotvall J. Bronchodilators. In: O'Byrne P, Thomson N, editors. Manual of asthma management. 2nd ed. London: WB. Saunders; 2001. p. 237-60. 4. Pedersen S. Management of acute asthma in children. In: O'Byrne P Thomson N, editors. Manual of asthma management. 2nd ed. London: W.B. Saunders; 2001. p. 237-60. 5. Liu A, Spahn J, Leung D. Childhood asthma. In: Behrman 6. Pederscn S, Bisgaard H. Clinical pharmacology and therapeutics. In: Silverman M, editor. Childhood asthma and other wheezing disorders. 2nd ed. London: Arnold; 2002. p. 247-76. 7. Restrepo RD. Use of inhaled anticholinergic agents in obstructive airway disease. Respir Care. 2007;52:833-51. 8. Qureshi F, Pestian J, Davis P, Zaritsky A. Effect of nebulized ipratropium on the hospitalization rates of children with asthma. N Engl J Med. 1998;339:1030-5. 9. Schuh S, Johnson DW, Callahan S, Canny G, Levison H. Efficacy of frequent nebulized ipratropium bromide added to frequent high-dose albuterol therapy in severe childhood asthma. J Pediatr. 1995;126:639-45.

10. Zorc JJ, Pusic MV, Ogborn CJ, Lebet R, Duggan AK. Ipratropium bromide added to asthma treatment in the pediatric emergency department. Pediatrics. 1999; 103:748-52. 11. Rodrigo G], Castro-Rodriguez J A. Anticholinergics in the treatment of children and adults with acute asthma: a systematic review with meta-analysis. Thorax. 2005;60:740-6. 12. Storr J, Lenney W Nebulised ipratropium and salbutamol in asthma. Arch Dis Child. 1986;61:602-3. 13. Ducharme FM, Davis GM. Randomized controlled trial of ipratropium bromide and frequent low dose"; of salbutamol in the management of mild and moderate acute pediatric asthma. J Pediatr. 1998;133:479-85. 14. Kartininingsih L, Setiawati L, Makmuri M. Comparison of clinical efficacy and safety between salbutamol-ipratropium :bulization compared to

salbutamol alone in asthma attacks bromide nebulization and salbutamol alone in children with asthmatic attack. Paediatr Indones. 2006;46:241-5. 15. Carruthers DM, Harrison BD. Arterial blood gas analysis or oxygen saturation in the assessment of acute asthma/ Thorax. 1995;50:186-8. 16. Rayner RJ, Cartlidge PH, Upton CJ. Salbutamol and ipratropium in acute asthma. Arch Dis Child. 1987;62:840-1. 17. Kumaratne M, Gunawardane G. Addition of ipratropium to nebulized albuterol in children with acute asthma presenting to a pediatric office. Clin Pediatr (Phita). 2003;42:127-32. 18. Sharma A, Madaan A. Nebulized salbutamol vs salbutamol and ipratropium combination in asthma. Indian J Pediatr. 2004;71:121-4. 19. Bateman ED, Hurd SS, Barnes PJ, Bousquet J, Drazen JM, FitzGeraid M, et al. Global strategy for asthma management and prevention: GINA executive summary. Eur Respir J. 2008;31:143-78. 20. Qureshi F, Zaritsky A, Lakkis H. Efficacy of nebulized ipratropium in severely asthmatic children. Ann Emerg Med. 1997;29:205-11. 21. Rodriguez-Roisin R. Acute severe asthma: pathopaysiology and pathobiology of gas exchange abnormalities. Eur Respir J. 1997;10:1359-71.

22. Rodrigo GJ, Rodriquez Verde M, Peregalli V, Rodrigo C. Effects of shortterm 28% and 100% oxygen on PaC02 and peak expiratory flow rate in acute asthma: a randomized trial Chest. 2003;124:1312-7. 23. Meert KL, Clark J, Sarnaik AE Metabolic acidosis as an underlying mechanism of respiratory distress in children with severe acute asthma. Pediatr Crit Care Med. 2007;8:519-23.

EFFICACY OF SYNBIOTIC AND PROBIOTIC TREATMENTS ON ACUTE WATERY DIARRHEA IN CHILDREN Ani Isti Rokhmawati, Wahyu Damayanti, Madarina Julia

Abstract Background In developing countries, acute watery diarrhea is a common cause of morbidity and mortality in children. Giving synbiotics or probiotics may decrease the severity of diarrhea. Objective To compare the efficacy of synbiotics and probiotics in decreasing the frequency of diarrhea, shortening the duration, and increasing patient body weight. Methods This was a double-blind, randomized clinical trial.to compare the effects of synbiotic vs probiotic treatment in children aged 6-59 months with acute watery diarrhea. This study was performed from October to December 2010 in two hospitals in Central Java. Subjects received either synbiotics or probiotics twice daily for five days. The measured outcomes were duration of diarrhea, daily frequency of diarrhea, and increase in body weight. Results There was no significant difference in the mean duration of the diarrhea in the synbiotic and probiotic groups, 3.92 days (SD 0.79) vs 3.80 days (SD 0.82), (P=0.35), respectively. Nor did we observe a significant difference in the mean increase in body weight in the synbiotic and probiotic groups, 150 g (SD 49.7) vs 160 g (SD 48.9), (P= 0.67), respectively. Conclusion We observed no significant differences in-efficacy of synbiotic and probiotic treatment for management of acute watery diarrhea. [Paediatr Indones. 2012;52:209-12]. Keywords: Acute watery diarrhea, synbiotic, probiotic

Acute watery diarrhea is a common cause of morbidity and mortality in children. Dehydration, as its main complication, causes the death of 5 to 10 million children in the world annually. The main treatments for managing acute watery diarrhea are rehydration, prevention of further dehydration, and dietetic treatment. The goal of dietetic treatment is to improve the microbial ecosystem of

the gut. Effects of probiotics in the gastrointestinal tract are improvement of lactose absorption, normalization of gut microflora, clearance of pathological microorganisms, and improvement cf humoral immunity by increasing IgA secretion. Intake of probiotics (living microorganisms), and synbiotics (consisting of a mixture of living microorganisms and oligosaccharides) has been demonstrated to modify the composition of the gut microflora, restore the microbial balance, hence, providing potential health benefits. Previous studies have assessed the efficacy of probiotics compared to placebo, as well as synbiotics compared to placebo, in the management of acute watery diarrhea. Both reported significant benefits. However, there has been little data on which of the two, probiotics or synbiotics, is more effective for treating acute watery diarrhea in children. We conducted this study to compare the effects of synbiotics and probiotics in decreasing the frequency of diarrhea, shortening the duration of diarrhea, and increasing body weight during illness. Methods We conducted this study in the pediatric ward of Soeradji Tirtonegoro Hospital, Klaten, and Muntilan Hospital, Magelang, from October to December 2010. The study design was a randomized, double-blind clinical trial. Subjects were children aged 6-59 months with acute water/ diarrhea admitted at the two hospitals. Acute watery diarrhea was defined as watery stools occurring more than three times per day and lasting for at least seven days. We excluded subjects with bloody diarrhea and diarrhea with severe complications, such as severe dehydration, metabolic acidosis, or seizures. Parents of subjects provided written informed consent. This study was approved by the Health and Medical Ethics Committee of the Gadjah Mada University Medical Faculty. The required number of subjects (176) was determined by hypothesis test, based on the means of the two populations. Subjects were consecutively allocated into the two treatment groups using a random number table. Only the appointed pharmacist had access to the subjects' allocated intervention, while subjects and physicians were blinded to the information (see study profile in Figure 1).

Before the study, subjects were examined to determine their level of dehydration, based on WHO guidelines. Subjects suffering from dehydration were rehydrated before weighing. The synbiotic and piobiotic sachets were administered twice daily for five days by nurses. Patient compliance was recorded. The physician evaluated the level of dehydration and frequency of diarrhea every two hours in the first six hours after admission, followed by evaluation every twelve hours. We evaluated the duration of diarrhea from acute watery diarrhea, as well as the increase in body weight during illness. Duration of diarrhea was defined as the time taken for diarrheal frequency tc decrease to less than three times per day. If subjects were discharged in less than five days, we asked parents to have their children weighed at the outpatient clinic on day five after admission. Figure 1. Study profile Results Of the 207 children recruited, 176 subjects completed the study. There were 88 subjects in each of the two groups. Baseline characteristics of subjects are shown in Table 1. The results of this study are shown in Table 2. There was no significant difference in the duration of the diarrhea (P~ 0.35), the daily frequency of diarrhea, or the increase in body weight (P=0.67). We observed potential side effects of giving synbiotics and probiotics, like bloating and flatulence. Three patients suffered from bloating in synbiotic group. No other gastrointestinal adverse effects were observed during the study.

Discussion We conducted this study to compare the effects of synbiotics and probiotics in decreasing the frequency of diarrhea, shortening the duration of diarrhea, and increasing body weight during illness. We found no significant differences between the two groups for mean daily frequency of diarrhea, mean duration of diarrhea or mean increase in body weight.

We observed side effects, such as bloating and flatulence, from both synbiotic and probiotic treatment. Three patients in the synbiotic group suffered from bloating. Fermentation of the substrate (prebiotics) produces hydrogen, which may cause bloating, flatulence, and diarrhea. But it is difficult to differentiate if these effects ate due the diarrhea itself or due to the additional substrates (prebiotics).

Table 1. Baseline characteristics of subjects Characteristics Characteristics Mean age, months SD Gender Males, n (%) Females, n (%) Socio-economic status1 Low, n (%) High, n (%) Nutritional status2 Undernourished, n (%) Well-nourished, n (%) 31 (35) 57 (65) 34 (38) 54 (52) 21 (24) 67 (76) 41 (76) 47 (34) 51 (58) 37 (42) 41 (47) 47 (53) Synbiotic group (n=88) 27.69 13.73 Probiotic group (n=88) 23.69 14.74

1 Determination of socio-economic status was based on the ratio of meal expenditure to monthly family income. In Central Java in 2009, high socioeconomic status was defined as a ratio of < 51.8%, while low socio-economic status was defined as a ratio of > 51.8%. 2 Nutritional status was categorized by body weight for height z-score, based on the WHO 2006 growth standard. Obesity was defined as a 7-score of > 2 SD; well-nourished was defined as a z-score of -2 SD to +2 SD; wasting was defined as a z-score of -2 SD to -3 SD; while severe wasting was defined as a z score of < -3 SD. These four categories were combined into 2 groups: well-nourished for obesity and well-nourished combined, and under-nourished for wasting and severe wasting combined.

Table 2. Results of the study

Variables Synbiotic group (n=88) Mean duration of diarrhea, 3.92 0.79 days + SD Mean frequency of diarrhea, per day + SD day 1 day 2 day 3 day 4 day 5 Mean increase in body weight, grams + SD 6.87 1.62 4.47 1.82 2.57 1.52 0.94 1.09 0.23 + 0.47 150 + 49.7 6.67 1.52 4.28 1.78 2.52 1.56 1.13 + 1.50 0.20 0.45 165 48.9 -0.20 -0.19 -0.05 0.19 -0.03 15.0 -0.67 to 0.26 -0.72 to 0.34 -0.51 to 0.4 -0.19 to 0.58 -0.17 to 0.10 -1.01 to 29.9 0.39 0.47 0.80 0.33 0.63 0.67 Probiotic Mean 95% CI P value group (n=88) Difference 3.80 0.82 -0.11 -0.35 to 0.12 0.35

Previous in vitro studies reported that synbiotics decreased the frequency of diarrhea, shortened the duration of diarrhea, and increased body weight better than probiotics. The addition of substrate for bacterial growth to the live microbes increases bacterial survival. Galactooligosaccharides were not only beneficial for the administered bacteria, but also for the residing bifidobacteria and lactobacilli in the intestines. Many species of bifidobacteria and lactobacilli other than the administered probiotics were observed in the feces after synbiotic treatment. Previous studies assessing the efficacy of probiotics compared to placebo, as well as synbiotics compared to placebo, in the management of acute watery diarrhea, reported significant benefits. This study is the first to compare administered synbiotics and probiotics in the management of acute watery diarrhea in children. However, we observed no beneficial effect of adding prebiotics into the probiotics, i.e the synbiotics, compared to the probiotics alone. If the substrate for bacterial growth was already present in the intestines, adding additional substrate might be useless. A limitation of our study was that we did not analyze food intake from daily diet. Food intake would affect the increase in body weight. Also, we

assessed the degree of dehydration based on clinical findings alone, not by checking urine specific gravity. In conclusion, we found no significant differences in daily frequency and duration of diarrhea, or increase in body weight between synbiotic and probiotic treatment in children with acute watery diarrhea. References 1. King CK, Gilass P., Bresee JS, Duggan C. Managing acute gastroenteritis among children, oral rehydration, maintenance, and nutritional therapy. c2003 [cited 2010 Sept]. Available from: http://www.cdc.gov/mmwr/previev/ mmwrhtml. 2. Branski D, Wilschanski M. Probiotics in gastrointestinal disorders. In: Behrman RE, Kliegman RM, Jenson HB. Nelson textbook of pediatrics. 18th ed. Philadelphia: Saunders; 2007. p. 1618-20. 3. Salazar L, Miranda L, Campos S, Chae W Lactobacillus casei strain GG in the treatment of infants with acute watery diarrhea: a randomized, double-blind, placebo-controlled clinical trial. BMJ Pediatr. 2004;4:1-6. 4. Cohen MB. Evaluation and treatment of the child with acute diarrhea. In: Rudolph CD, Rudolph AM, Hostetter MK, Lister G, Siegel NJ. editors. Rudolph's Pediatrics. 21st ed. New York: McGraw-Hill Medical Publishing Division; 2003. p. 656-60. 5. Garleb KA, Snowden MK, Wolf BW, Chow J. Aplication of

fructooligosaccharides to medical foods as fermentable dietary fiber. Nutr J. 2002;2:43-5. 6. Kullen MJ, Amman MM, Shaughnessy MJ, O'Sullivan DJ, Busta FF, Brady LJ. Differentiation of ingested and endogenous bifidobacteria by DNA fingerprinting demonstrates the survival of an unmodified strain in the gastrointestinal tract of human, j Nutr. 1997;127:89-94. 7. Hickson M, Aloysius LD, Muthu N, Rogers T, Want S, Rajkumar C. Use of probiotic Lactobacillus preparation to prevent diarrhea associated with antibiotics: 2007;5:11-5. randomized, double-blind, placebo-controlled trial. BMJ.

8. Rumessen J, Gudmand HE. Fructans of chicory: intestinal transport and fermentation of different chain lengths and relation to fructose and sorbitol malabsorption. Am Clin Nutr J. 1998;68:337-64. 9. Badan Pusat Statistik. Pengeluaran dan Konsumsi Penduduk per Provinsi, Pusat Statistik Indonesia 2008/2009. c2010 [cited 2010 Sept]. Available from: http://www.bps.gn.id 10. Dughera L, Flia C, Navino M, Cisaro F, the ARMONIA study group. Effects of symbiotic preparations on constipated irritable bowel syndrome symptoms. Acta Biomed. 2007;78:111-6. 11. Rautawa S, Salminem S, Isolauri E. Functional food science gastrointestinal physiology and function. Br ] Nutr. 2009;101:1722-6. 12. Schezenmeir J, Heller K, McCue M, Liamas C, Lam W Burow H, duration of diarrhea. Benefit of oral supplementation with and without synbiotics in young children with acute bacterial infection. Clin Pediatric. 2004;43:Z39-49. 13. Simakachorn N, Pichaipat V, Rithipornpaisam V, Kongkaew C, Tongpradit P Varavithya W. Clinical evaluation of the addition of lyophilized, heat-killed Lactobacillus acidophilus LB to oral rehydration therapy in the treatment of acute diarrhea in children. ] Pediatr Gastroenterol Nutr. 2000;30:68-72.

THE EFFECT OF EXCLUSIVE BREASTFEEDING ON REDUCING ACUTE RESPIRATORY INFECTIONS IN LOW BIRTH WEIGHT INFANTS Karolina Trigemayanti Tallo1, I Kompiyang Gede Suandi1, Setya Wandita2

Abstract Background Infants who are breastfed receive natural protection against certain infections. Despite the known benefits of exclusive breastfeeding, many Indonesian mothers choose to supplement with formula. There have been few Indonesian s:udies on the effect of exclusive breastfeeding in reducing acute respiratory infections in low birth weight infants in their first four months of life. Objective To investigate the effect of exclusive breastfeeding in reducing the incidence of acute respiratory infections in low birth weight infants during their first four months of life. Methods We conducted a prospective cohort study on low birth weight babies in Sanglah Hospital, Denpasar. The total number of subjects was 181. The incidence of acute respiratory infections during the first 4 months of life and the duration of breastfeeding were assessed by questionnaires. Data was analyzed with Chi square and logistic regression tests. Results Infants who were exclusively breastfed for 4 months had a lower risk of acute respiratory infection than those who were not exclusively breastfed (RR 0.07; 95% CI 0.03 to 0.14; P=0.001). After adjustment for gestational age, parity, maternal nutritional status, family size, smoke exposure, and history of atopy, infants who were exclusively breastfed still had a lower risk for acute respiratory infection than those who were not exclusively breastfed (RR 0.06; 95% CI 0.03 to 0.13; P = 0.001). Conclusion Exclusive breastfeeding reduced the risk of acute respiratory infection in low birth weight infants in the first four months after birth. [Paediatr Indones. 2012;52:229-32]. Keywords: low birth weight, breastfeeding, acute respiratory infection

Birth weight is a strong predictor for growth and survival of infants. Low birth weight babies have specific problems caused by immaturity of multiple organ systems. Therefore, low birth weight babies have a higher risk of infection, especially respirator/ infections. Acute respiratory infection is a cause of high infant mortality and morbidity in Indonesia (24-25%). Breastfeeding practices differ among Indonesians. Based on data from the 2002 Indonesian Demographic Health Survey (SKDI - Survei Demografi Kesehatan Indonesia), only 3.7% of newborns received breast milk on the first day of their lives. Data from the Ministry of Health reported the exclusive breastfeeding rate in 2008 to be 37.98% in Bali. Breast milk can protect infants from certain infections, such as respiratory infection, gastrointestinal infection, and sepsis. We aimed to determine the effect of exclusive breastfeeding on reducing acute respiratory infections in low birth weight infants during the first four months of life. Methods This prospective cohort study was performed in Sanglah Hospital, Denpasar, from January 2010 to April 2011. Subjects were low birth weight infants born in the hospital, with birth weight of 2,000 -2,500 grams and gestational age > 34 weeks. They were singletons and received breast milk or breast milk and formula combination either directly or by bottle, spoon, or nasogastric catheter. Subjects who got breastmilk combined with formula milk and/or additional food or only formula milk were classified as not exclusively breastfed. Subjects' parents provided written informed consent. Subjects were excluded if they suffered from complications during labor, infections during their time in the hospital nursery, or had contraindications for breastfeeding or congenital anomalies. The required number of subjects was 181, based on a=5% and power 80%. Subjects were recruited by consecutive sampling and assigned to either the exclusively breastfed group or the non-exclusively breastfed group At the time of hospital discharge, mothers were given diaries to record their breastfeeding practices and their child's illnesses. They were asked to visit the Outpatient Clinic of the Department of Child Health, Sanglah Hospital,

Denpasar, on a monthly basis. During each visit, mothers filled questionnaires on the episodes of acute respiratory infections in their child, as well as the continuity of their breastfeeding practice. Parents were requested to take their infants immediately to the Pediatric Outpatient Clinic if their infants were ill. Sick infants were then examined by eligible pediatric residents. If the parents and infants were unable to attend the monthly clinic visit, a research assistant would call or make a home visit on a suggested day. Questionnaires would then be filled at the subject's residence. We used Chi square and logistic regression tests to analyze the data. A P value of < 0.05 was considered to be statistically significant. This study was approved by the Ethics Committee of the Udayana University Medical School/ Sanglah Hospital. Denpasar. Table 1. Baseline characteristics of subjects

Characteristics months 4 (n = 91)

Characteristics Exclusively breastfed for 4 months (n = 90) Not exclusively breastfed for 4 months (n = 91)

Infant characteristics Male gender, n (%) Small for gestational age, n (%) Family characteristics Mean maternal age, years (SD) Parity, n (%) First parity Multiparity Maternal nutritional status, n (%) Undernourished Well-nourished Overweight Obese 17(19) 67 (74) 6(7) 0(0) 19(21) 65(71) 6(7) 1(1) 51 (57) 39 (43) 41 (45) 50 (55) 25.57(4.71) 26.02 (4.75) 46(51) 43 (48) 44 (48) 36 (39)

Mean monthly family income, Rupiah (SD) * Family size, n (%) < 4 people Combination contraception, n (%) Yes Smoke exposure, n (%) Yes Ethnicity, n (%) Balinese Javanese Others DPT immunization, n (%) Yes Family history of atopy, n (%) Yes

1,390,000 (688,464)

1,350,000 (580,684)

51 (57)

39 (43)

1(1)

12(13)

55(61)

66 (73)

63 (70) 19(21) 8(9)

61 (67) 23 (25) 7(8)

87 (97)

91 (100)

32 (35) .

38 (42)

Results One hundred eighty-six subjects fulfilled our inclusion criteria. Three subjects were lost to follow up and two dropped out the study due to change of residence and death. Hence, 181 subjects completed the study. The baseline characteristics for each group are shown in Table 1. In unadjusted analysis, infants who were exclusively breastfed for four months were at lower risk for having acute respiratory infection than those who were not exclusively breastfed for four months (RR 0.07; 95% CI 0.03 to 0.14) (Table 2). Table 3 shows the risk of acute respiratory infection after adjustments for gestational age, parity, maternal nutritional status, family size, smoke exposure, and family history of atopy. We found that infants who were exclusively breastfed for four months had a lower risk of acute respiratory infection compared to those who were not exclusively breastfed for four months (Adjusted RR 0.06; 95% CI 0.03 to 0.13). Table 2. Risk of acute respiratory infection (ARI)
Incidence of ARI RR 95% Cl P value

Exclusively breastfed (n= 90 ) Not exclusively breastfed (n=91)

23 76

0.07

0.033 to 0.140

0.0001

Table 3. RR for ARI adjusted for size for gestational age, parity, maternal nutritional status, family size, smoke exposure, and history of atopy
Adjusted RR Exclusively breastfsd Size for gestational age Small gestational age Parity First parity Maternal nutritional status Under-nourished Familv size < 4 people Smoke exposure Yes Family history of atopy Yes 0.96 0.45 to 2.07 0.923 1.14 0.52 to 2.50 0.739 2.01 0.79 to 5.10 0.140 1.60 0.79 to 3.22 0.186 0.59 0.23 to 1.46 0.254 0.84 0.39 to 1.79 0.649 0.06 95% CI 0.03 toy. 13 P value 0.001

Discussion There have been few Indonesian studies on the relationship between exclusive breastfeeding and acute respiratory infection, particularly in low birth weight infants. Most studies used normal weight, newborn infants as subjects. In previous studies, infants given formula milk for four months were reportedly at greater risk for acute respiratory infections compared to those who were exclusively breastfed. Another study reported that shorter duration of breastfeeding increased the risk of acute respiratory infections. We found that exclusively breastfed low birth weight infants had a lower risk of acute respiratory infection compared to those who were not exclusively breastfed for the first four months of life. It is thought that bioactive components in breast milk protect against acute respiratory infections. Passive protection from breast milk affects the immune

system's response by various mechanisms, such as the immune system's maturation, anti-inflammation, immunomodulation, and antimicrobial activity. Immune components found in breast milk, such as secretory

immunoglobulin A (slgA) and interferon, protect low birth weight infants from infection. Also, skin to skin contact during breastfeeding stimulates the production of specific antibodies from mothers against infection. Secretory Ig A is- one of three main classes of immunoglobulins found in colostrum and breast milk Secretory Ig A concentration in breast milk of mothers with low birth weight infants is higher than in mothers with normal birth weight infants.' Secretory Ig A can activate the complement system through alternative pathway and in cooperation with macrophages can also phagocytize various microorganisms. Secretory Ig A also plays an important role in the defense against syncytial virus, and macroglobulin-like substance can inhibit influenza and parainfluenza viruses. The relatively short duration of follow-up was a limitation of this study. The effects of exclusive breastfeeding were studied only until the infants were four months of age. The recommended minimum duration of exclusive breastfeeding for low birth weight infants is four months. Further research with a longer duration of breastfeeding is needed to determine the long-term effect of breastfeeding on acute respiratory infections. In conclusion, in low birth weight infants, subjects who were exclusively breastfed for the first four months of life had reduced risk of acute respiratory infection compared to subjects who were not exclusively breastfed. Acknowledgment We would like to express our highest gratitude to I Gde Raka Widiana, MD for his help in constructing the methodology and statistical analyses in this study. References 1. Pojda J, Kelley L. Low Birth Weight. ACC/SCN Nutrition Policy Paper. 2000;2:18-28.

2. Indonesian Ministry of Health. Peningkatan pemberian ASI sampai tahun 2005. In: Utoro R. editor. Strategi nasional program for appropriate technology in health. 1st ed. Jakarta: Depkes; 2005. p. 1-12. 3. Dinas Kesehatan Propinsi Bali. Data cakupan ASI tahun 2008. Dinas kesehatan Propinsi Bali: Denpasar; 2008. 4. Raisler J, Alexander C, O'Campo E Breastfeeding and infant illness: a doseresponse relationship? Am J Public Health. 1999;89:25-30. 5. Chantry CJ, Howard CR, Auinger E Full breastfeeding duration and associated decrease in respiratory tract infection in US children. Pediatrics. 2006;117:425-32. 6. Bachrach VRG, Schwarz E, Bachrach LR. Breastfeeding and the risk of hospitalization for respiratory disease in infancy. Arch Pediatr Adolesc Med. 2003;157:237-43. 7. Duitjs L, Jaddoe V, Hofman A, Moll H. Prolonged and exclusive breastfeeding reduces the risk of infectious disease in infancy. Pediatrics. 2010;126:18-25. 8. Alarcon ML, Villapando S, Arturo F. Breastfeeding lowers the frequency and duration of acute respiratory infection and diarrhea in infants under six months of age. J Nutri 1996;127:436-42. 9. Maria AQ, Kelly Y], Amanda S. Breastfeeding hospitalization for diarrheal and respiratory infection the United Kingdom millennium cohort study. Pediatrics? 2007;119:837-42. 10. Kelly DC. Early nutrition and the development of immune function in the neonate. Proc Nutr Soc. 2000;59:177-85. 11. Gross SJ, Buckley RH, Wakil SS, McAllister DC, David RJ, Faix RG. Elevated IgA concentration in milk produced by mothers delivered ot preterm infants. J Pediatr. 1981;99.389-93. 12. Ryan E Kawaoka Y. cc2-macroglobulinis the major neutralizing inhibitor of influenza. Am J Epidemiol. 1993;119:516-25.

THE RELATIONSHIP BETWEEN PLEURAL EFFUSION INDEX AND MORTALITY IN CHILDREN WITH DENGUE SHOCK SYNDROME Novrianti Hawarini1, Muhammad Sholeh Kosim1, M Supriatna1, Yusrina Istanti1, Eddy Sudijanto2

Abstract Background Dengue shock syndrome (DSS) mortality rate is still high. The extenf of plasma effusion in dengue shock syndrome can be identified in the right lateral decubitus position on chest x-ray, and quantified by the pleural effusion index (PEI). It is thought that PEI value can be used to predict DSS mortality in children. Pleural effusion in DSS patients can cause respiratory failure and death. Objective To determine the relationship between PEI'and ' mortality in children with DSS. Methods This cross-sectional, retrospective study was held in the Dr. Kariadi Hospital, Semarang, Indonesia. Data was taken from medical records of pediatric intensive care unit (PICU) patients with DSS from January 2009 to January 2011. DSS diagnosis was confirmed by clinical and radiological manifestations. PEI diagnosis was established by the presence of tluid in the pleural cavity on pulmonary radiological examinations. X-rays were interpreted by the radiologist on duty at the time. Chi square and logistic regression tests were used to analyze the data. Results There were 48 subjects with DSS, consisting of 18 males (37.5 %), and 30 females (62.5%). Twenty-nine subjects (60.4%) survived and 19 (39.6%) died. One patient (2.1%) had.PEI <6%, 4 (8.3%) had PEI 6-i5%, 17 (35.4%) had PEI 15-30%, and 26 (54.2%) had PEI > 30% on their x-rays The mortality rate of DSS with PEI 15-30% was 11.8% (95% CI 0.021 to 0.564; P<0.005) and PEI >30% was 65.4 % (95% CI 3,581 to 99,642; P < 0.005). Conclusion PEI > 15% was a risk factor for mortality in children with DSS. [Paediatr Indones. 2012;52:239-42], Keywords: pleural effusion index, mortality in dengue shock syndrome

Dengue hemorrhagic fever is an important cause of morbidity in Asian children, and dengue shock syndrome causes a significant number of childhood deaths. DSS is characterized by a massive increase in systemic capillary permeability with consequent hypovolemia. WHO defines DSS as DHF plus signs of circulatory failure manifested by rapid and weak pulse, narrow pulse pressure (< 20 mmHg) or hypotension for age, prolonged capillary refill, cold and clammy skin and restlessness. Onset of shock is acute and occurs at the time of defervescence, usually after 2-5 days of fever. During shock, patients may have subnormal body temperature, cold and clammy skin, as well as rapid and feeble pulse. Pleural effusion and ascites measurements may be used to predict the development of DSS. Pleural effusion occurs in the phase of plasma leakage, causing decreased chest compliance and reduced functional residual capacity. Furdiermore, pleural effusion causes hypoxemia and increased breathing effort. Clinical manifestations caused by pleural effusion depend upon the volume of pleural fluid, in addition to lung parenchymal pathology, such as acute respiratory distress syndrome.6 Right lateral decubitus position on chest x-ray is used to evaluate pleural effusion. The degree of plasma leakage may be quantified by the PEL PEI is calculated to be 100% times the maximum width of the right pleural effusion, divided by the maximal width of the right hemithorax.7 The objective of this study was to evaluate the use of PEI values to predict mortality in DSS in children. Methods The retrospective, cross-sectional study was held in the Dr. Kariadi Hospital, Semarang, Indonesia from January 2009 to January 2011. Data was taken from medical records of children with DSS in the PJCU who fulfilled the inclusion criteria. Eligible subjects were children aged 1-14 years with diagnoses of dengue syndrome (DS) according to WHO criteria, and were not having septic or shock condition. DSS diagnosis was confirmed by clinical and radiological

manifestations. PEI was assessed from lateral decubitus position on chest x-rays and calculated by the formula A/B x 100% (Figure l). Radiological examination results were reviewed by the on-duty radiologist. We analyzed data using Chi square and logistic regression analyses with SPSS software version 17.0. Results There were 48 PICU cases of DSS from January 2009 to Januari 2011, consisting of 18 males (37.5%) and 30 females (62.5%). Twenty-nine subjects survived (60.4%), and 19 died (39.6%), as shown in Table 1. Table 2 shows the PEI groupings based on x-ray findings. The relationship of PEI to death was observed in PEI values of greater than 15%, with a statistically significant association in the two highest PEI categories, 15-30% and > 30% (Table 3). Discussion This study was conducted to determine the relationship of plasma leakage severity, as measured by PEI, to mortality in DSS patients. Subjects' genders in our study were 37.5% male and 62.5% female. A 1987 Singaporean study reported a higher number of cases of men than women with a ratio of 1.9 : 1, while a 1993 That study reported girls to be two times more frequently hospitalized due to dengue. In a 1990 Indonesian study, cited from Supriatna MS, there was no significant difference between males and females in the number of DHF cases and shock events. From a total of 48 DSS patients, 19 died (39.6%) and 29 lived (60.4%). Nationally, DHF mortality rate was reported to be low (2.5% in 1997) and remains to be below 3%.13 In Semarang in 2004, there were 1621 dengue cases with an incidence rate of 11.8 per 10,000 population and a case fatality rate of 0.43%. DSS mortality in the Dr. Kariadi Hospital PICU decreased from 12% in 2002 to 10.8% in 2004. Figure 1. Pleural effusion index calculation8 Table 1. Characteristics of subjects Characteristics n = 48 %

Sex Male Female Mortality status Survived Died 29 19 60.4 39.6 18 30 37.5 62.5

DSS is defined as DHF with signs of circulatory failure, including narrow pulse pressure (20 mm Hg), hypotension, or frank shock. The prognosis in DHF/ DSS depends on prevention or early recognition and treatment for shock. In hospitals with experience in treating DSS, the case fatality rate in DHF may be as low as 0.2%. Once shock has set in, the fatality rate may be much higher (12% to 44%).M Table 2. PEI groupings based on chest x-ray findings PEI <6% 6-15% 15-30% >30% N = 48 1 4 17 26 % 2.1 8.3 35.4 54.2

Table 3. Relationship of mortality rates to PEI PEI 15-30% >30% Death n (%) 2(11.8) 17(65.4) OR 0.110 18,889 95% CI 0.021 to 0.564 3,581 to 99,642 P <0.005 <0.005

The presence of plasma leakage can be observed by the presence of pleural effusion, hemoconcentration, and hypoalbuminemia. In some studies, significant pleural effusion has been associated with shock and mortality. In our study, DSS patients had varying PEI measurements: PEI <6 % (2.1%), PEI 6 15% (8.3%), PEI 15-30% (35.4%), and PEI >30% (54-2%). In a Thai study in DSS patients, pleural effusion was found in 22 of 26 DSS cases with an average of PEI of 4-1%.16 The average PEI from our hospital in 2004 was 18.29% (DSS) and 4.75% (non-DSS).

A limitation of our study was that we could not use a Kappa test for radiologists' assessment-of x-ray findings, since this was a retrospective study. Our study revealed a significant relationship between mortality rate from DSS and PEI. In conclusion, PEI > 15% was a risk factor for mortality in children with DSS. References 1. Ngo NT, Cao XT, Kneen R, Wills B, Nguyen VM, Nguyen TQ, et al. Acute management of dengue shock syndrome: a randomized double-blind comparison of 4 intravenous fluid regimens in the first hour. Clin Infect Dis. 2001;32:204-13. 2. Soegijanto S, Budivanto, Kartika, Taufik, Amor. Update management of dengue complication in pediatric. Indonesian J Trop Infect Dis. 2011;2:1-11. 3. Catharina S, Tatty E. S, Eric CM, Robert J D, et al. Risk factors for mortality in dengue shock syndrome (DSS). Media Medika Indonesiana. 2009;43:213-9. 4. Agaral R, Kapoor S, Nagar R, Misra A, Tandon R, Mathur A, et al. A clinical study of the patients with dengue hemorrhagic fever during the epidemic of 1996 at Lucknow, India. Southeast Asian J Trop Med Public Health. 1999;30:735 40. 5. World Health Organization (WHO) Regional Office for South-East Asia. Guidelines for treatment of dengue fever/ dengue hemorrhagic fever in small hospitals. New Delhi: WHO Regional Officer for South East Asia; 1999. 6. Alkrinawi S, Chernick V. Pleural fluid in hospitalized pediatric patients. Clin Pediatr. 1996;35:5-9. 7. Syahrial R, Sukonco K, Iwan E. Radiologi diagnostik. Jakarta: GayaBaru; 1998. p. 115-8. 8. Sathupan P Khongphattanayothin A, Srisai J, Srikaew K. The role of vascular endothelial growth factor leading to vascular leakage in children wi4i dengue virus infection. Ann Trop Paediatr. 2007;27:179-84. 9. Vaughn DW Green S, Kalayanarooj S, Innis BL, Nimmannitya S, Suntayakorn S, et al. Dengue in the early febrile phase: viremia and antibody responses. J Infect Dis. 1997;176:322-30.

10. Committee of Epidemic Diseases. Surveillance for dengue fever/dengue hemorrhagic fever in Singapore. Epidemiological News Bulletin (Singapore). 2002;28:25-30. 11. World Health Organization (WHO). Guidelines for case reporting and management. Dengue fever and fengue hemorrhagic fever. New Delhi: WHO Regional Office for South-East Asia; 1999. 12. Supriatna MS. Perbedaan gangguan fungsi hati pada demam berdarah dengue [master's thesis]. [Semarang]: Diponegoro University; 2004. 13. Setiati TE. Pengelolaan syok pada demam berdarah dengue anak. In: Sutaryo.Hagung P Mulatsih S, editors. Tatalaksana syok dan perdarahan pada demam berdarah dengue. Yogyakarta: Medika FK UGM; 2004. p. 75-86. 14. Rigau-Perez JG, Clark GG, Guhler DJ, Reiter R Sander EJ, Vorndam AV. Dengue and dengue haemorrhagic fever. Lancet. 1998;352:971-7. 15. Setiati TE. Faktor hemostasis dan faktor kebocoran vaskular sebagai faktor diskriminan untuk memprediksi syok pada DBD [dissertation]. [Semarang]: Diponegoro University; 2004. 16. Pramuljo HS. Peran pencirraan pada demam berdarah dengue. In: Harun SR, Safari HI, editors. Naskah lengkap pelatihan bag! pelatih dokter spesialis anak dan dokter spesialrs ptnyakit dalam. Jakarta: BP FKUI; 2000. p. 63-72.