Lysosomal enzymes in preterm infants with bronchopulmonary dysplasia: a potential diagnostic marker

Clinica Chimica Acta 278 (1998) 23–34

Lysosomal enzymes in preterm infants with bronchopulmonary dysplasia: a potential diagnostic marker Giancarlo Goi a , Chiara Bairati a , Luca Massaccesi a , Adriana Lombardo a , Luisa Bonafe` b , Vincenzo Zanardo b , b, Alberto Burlina * a

Department of Medical Chemistry and Biochemistry, Medical School, University of Milan, Milan, Italy b Department of Paediatrics, Medical School, University of Padua, Padua, Italy Received 13 May 1998; received in revised form 21 July 1998; accepted 7 August 1998

Abstract Some lysosomal glycohydrolases (N-acetyl-b-D-glucosaminidase and their major isoenzymes, b-D-glucuronidase, a -D-galactosidase, b-D-galactosidase and a-D-glucosidase) were investigated in the plasma of 36 preterm infants with respiratory distress, 11 of whom developed bronchopulmonary dysplasia (BPD), in order to evaluate the role of the lysosomal apparatus in the disease. Enzyme activity was assayed fluorimetrically; the major N-acetyl-b-D-glucosaminidase (NAG) isoenzymes were separated using a routine chromatofocusing procedure; the diagnostic efficiency was evaluated by Bayes theorem. The mean levels of almost all glycohydrolases considered were significantly higher in BPD than in non-BPD infants. Among NAG major isoenzymes, an increase was found only in form A. No variation was evident in the plasma levels of glycohydrolases during dexamethasone therapy. Data from a retrospective analysis performed in all preterms considered, show that a-D-galactosidase and b-D-galactosidase differentiate a posteriori BPD and non-BPD subjects. These enzymes, after a priori verification of their diagnostic potential in preterm infants at risk of BPD development, could acquire an important predictive value.  1998 Elsevier Science B.V. All rights reserved. Keywords: Lysosomal enzymes; Bronchopulmonary dysplasia; Preterm infants; Diagnostic marker; Dexamethasone therapy; N-Acetyl-b-D-glucosaminidase isoenzymes

*Corresponding child.pedi.unipd.it

author.

Tel.:

1 39-335-362-953;

fax:

1 39-49-821-3569;

0009-8981 / 98 / $ – see front matter  1998 Elsevier Science B.V. All rights reserved. PII: S0009-8981( 98 )00129-6

e-mail:

burlina@

24

G. Goi et al. / Clinica Chimica Acta 278 (1998) 23 – 34

1. Introduction Bronchopulmonary dysplasia (BPD), first described in 1967 in premature neonates on mechanical ventilation and oxygen supplementation for respiratory distress syndrome [1], is now the most common cause of chronic lung disease in infants, and its prevalence is rising [2] due to improved survival rates for extremely premature neonates. Current approaches to understanding the etiology of BPD, which is considered multifactorial [3,4], include the study of factors involved in immature lung injury and repair, including the mechanism underlying the inflammatory process [5]. Studies in animal models and in premature humans show that the activity of some lysosomal enzymes varies during the course of postnatal development [6–8]. One of these enzymes, N-acetyl-B-D-glucosaminidase, has been proposed as a marker for the early identification of necrotizing enterocolitis in preterm infants [9]. It has, moreover, been found that dexamethasone, a glycoactive steroid recently introduced in the treatment of BPD, can alter intra- and extracellular lysosomal enzyme levels [10–12]. A great deal of information is available on plasma lysosomal enzymes in different physiological conditions and in acquired diseases [13–17], also in newborns [18,19]. In the light of our findings for lysosomal enzymes and isoenzymes, including the standardisation of micromethods for their assay [20–22] and the evaluation of their behaviour, also in newborns [18,19], our aim was to investigate some lysosomal glycohydrolases in the plasma of preterm infants with respiratory distress, some of whom had BPD, in order to evaluate the role of the lysosomal apparatus in the disease and to ascertain whether one or more of the lysosomal glycohydrolases could be utilised in the early identification of subjects at risk of BPD. We also evaluated the behaviour of the enzymes in the course of prolonged dexamethasone treatment.

2. Materials and methods

2.1. Chemicals and other products Commercial chemicals employed were the purest available. The water routinely used was freshly redistilled in a glass apparatus. 4-Methylumbelliferone (4-MU), purchased from Fluka GmbH (Bucks, Switzerland) was recrystallized from ethanol three times; 4-MU-glycosides were purchased from Melford (Suffolk, UK). Polybuffer exchanger PBE-94 and polybuffer-74-HCl were from Pharmacia Fine Chemicals (Uppsala, Sweden). Dexamethasone was supplied by Decadron (RMSD, Haarlem, The Netherlands).

G. Goi et al. / Clinica Chimica Acta 278 (1998) 23 – 34

25

2.2. Subjects The study population consisted of a group of 36 preterm infants at high risk for bronchopulmonary dysplasia [23], prospectively selected among babies admitted to the Department of Pediatrics of Padua University. On the basis of positive pressure ventilation during the first 2 weeks of life for a minimum of 3 days and of clinical signs of abnormal respiratory function persisting longer than 28 days and requiring supplemental oxygen, in association with the presence of a chest radiograph with diffuse abnormal characteristic findings [24], BPD was diagnosed in 11 of these babies (six males and five females; gestational age: 27.862.5 weeks, range 24–31; birth weight: 9806189 g, range 640–1200). They were treated with dexamethasone (0.5 mg / kg / day for three days and 0.25 mg / kg / day for the next three days, followed by a 10% decrease every three days until 42 days of treatment were completed, as reported [24]). The remaining 25 premature infants (15 males and 10 females; gestational age: 30.961.6, range 28–33; birth weight: 13136221, range 870–1630) who did not develop BPD were used as a reference group. The neonatal age (weeks) at sampling times was 28.9611 (range: 10–59) and 30.3619 (range: 3–64) for preterm reference infants and BPD infants respectively. All data are expressed as mean6SD. At entry, none of the infants had clinical or laboratory signs of patent ductus arteriosus, active bacterial or viral infection, congenital heart disease or major congenital anomalies. Following routine guidelines of our nursery, nutritional and metabolic support were given and all subjects received a formula for premature infants ( , 150 ml / kg / die). Caffeine was routinely used for respiratory stimulation to facilitate breathing and to treat recurrent apnea and bradycardia. Diuretics and bronchodilatators were only occasionally used. When necessary, packed red cell transfusion was given to keep the capillary hematocrit above 40% for as long as infants required supplementary oxygen. Weight, parenteral and enteral intake were recorded daily. Routine biochemical controls were measured at least once a week, while blood sugar was monitored at least once a day. Blood samples for lysosomal enzyme assay were drawn from all patients during routine controls and in BPD patients also during dexamethasone administration (at 21 and 42 days of treatment). This protocol was approved by the local Committee for Bioethics.

2.3. Sample storage Plasma was prepared immediately after the blood was drawn, by centrifuging at 15 min at 3000 3 g blood (400 ml) treated with sodium citrate (final concentration: 11 mmol / l) known not to affect lysosomal enzymes activities [20]. Ethylene glycol was added (30%, v / v, final concentration) to the plasma

26

G. Goi et al. / Clinica Chimica Acta 278 (1998) 23 – 34

samples immediately after collection [21]. The samples were then stored at 2 208C until assayed.

2.4. Enzyme assay The following enzymes were studied: N-acetyl-b-D-glucosaminidase (NAG) (EC 3.2.1.30) and its major isoenzymes, b-D-glucuronidase (EC 3.2.1.31), a-D-galactosidase (EC 3.2.1.23), b-D-galactosidase (EC 3.2.1.22) and a-Dglucosidase (EC 3.2.1.20). Their activities were determined fluorimetrically, using 4-methyl-umbelliferyl-glycosides as substrates, following the indications of Lombardo et al. [20]. Briefly: in a final volume of 250 ml, the incubation mixture contained 25 ml of the appropriate 50 mmol / l buffer system at the appropriate pH, 125 ml of specific substrate aqueous solution, 100 ml of plasma appropriately diluted with saline solution or, in the case of NAG isoenzymes assay, 100 ml of the effluent from the column. The mixtures were incubated at 378C (30 min for all enzymes and 1 h for NAG isoenzymes). A stable liquid material was employed for calibration purposes [21]. The enzyme activities were expressed as mU / l of plasma.

2.5. Separation of N-acetyl-b -D-glucosaminidase ( NAG) isoenzymes NAG isoenzymes were separated using the routine chromatofocusing procedure described by Goi et al. [22]. The different isoenzyme fractions were eluted by gravity and the enzyme activity of the eluates was measured fluorimetrically. Briefly: undialysed plasma (100 ml) was applied to a microcolumn containing 350 ml of exchanger PBE-94 previously equilibrated with 25 mmol / l piperazine–HCl buffer, pH 5.5. After the loading and complete absorption of the sample, the first fraction, eluted with 10 ml of the same buffer, contained isoenzymes B 1 I 1 . Isoenzyme I 2 was then eluted with 7.0 ml of polybuffer 74, pH 4.8, followed by isoenzyme A eluted with 10 ml of the same buffer, pH 4.0. The entire procedure was carried out at room temperature. Before being used again, the column was washed with 5 ml of 1 mol / l NaCl solution and re-equilibrated with 10 ml of 25 mmol / l piperazine–HCl buffer, pH 5.5.

2.6. Statistical analysis Test of skewness and kurtosis did not show significant differences from normal distribution and parametric analysis techniques were used. Following Snedecor and Cochran’s indications [25], means were compared by one-way analysis of variance and correlation coefficients were calculated. Multiple

G. Goi et al. / Clinica Chimica Acta 278 (1998) 23 – 34

27

comparisons of means were made using the Bonferroni multiple test. The SPSS / PC package was utilised for all analyses [26]. The specificity, sensitivity and the diagnostic efficiency together with the posterior probability of positive or negative results were evaluated according to Bayes theorem [27], and the terms we used are defined as follows: Diagnostic specificity: the fraction of true-negative results out of all non-BPD preterm infants; diagnostic sensitivity: the fraction of true-positive results out of all BPD preterm infants; diagnostic efficiency: difference between sensitivity and 1-specificity; posterior probability of positive results (PPpos ): the fraction of true positives out of all positive results; posterior probability of negative results (PPneg ): the fraction of true negatives out of all negative results. Because data on enzymes were evaluated retrospectively, the term ‘posterior probability’ rather than ‘predictive value’ was utilised.

3. Results Table 1 reports the results of enzymatic assays performed on the plasma of 36 preterm infants with respiratory distress, 11 of whom had BPD and were not yet on dexamethasone therapy. The mean levels of the glycohydrolases considered, appear significantly higher in preterm BPD infants than in those without BPD. Among all enzymes considered, a-D-glucosidase was practically unaffected; b-D-glucuronidase, which appeared to be markedly increased in preterm BPD infants (nearly 3 fold), was the only enzyme demonstrated to have a positive correlation with neonatal age (P , 0.002, r 5 0.81). Among NAG isoenzymes, a significant increase was found only for the major form A. The single values of glycohydrolases in BPD and non-BPD infants are reported in Fig. 1, which shows that for at least two enzymes, a-D-galactosidase and b-D-galactosidase, the two preterm populations are more differentiated. Table 1 shows also the effects of dexamethasone administration. No significant variation was found in the plasma levels of glycohydrolases during drug therapy, except for b-D-glucuronidase, for which a positive correlation with age was also evidenced, as reported above. A retrospective analysis was performed in the total preterm population to make a preliminary assessment in order to evaluate whether the differences between glycohydrolases plasma levels in the two preterm infant groups could be utilised to distinguish between BPD and non-BPD infants irrespective of clinical parameters. For each enzyme the best diagnostic efficiency value was calculated and utilised for the identification of the discriminant. The relevant data are reported in Table 2. The diagnostic efficiency was minimal for a-D-glucosidase and maximal for a-D-galactosidase. In order to obtain a better differentiation between BPD and non-BPD infants,

28

Enzyme

Reference preterm infants (n 5 25)

B-D-Glucuronidase 173061720 b-D-Galactosidase 3576349 a-D-Galactosidase 150690 a-D-Glucosidase 170670 N-Acetyl-b-D21 20068830 glucosaminidase (NAG) NAG isoenzyme A 11 90065070 NAG isoenzymes B 1 I 1 797063915 NAG isoenzyme I 2 16356795

BPD infants before treatment (n 5 11)

BPD infants 21 days after treatment (n 5 11)

BPD infants 42 days after treatment (n 5 7)

521064390(***) 9336308(***) 265670(***) 200670 29 55069225(**)

899063830 10776423 3106180 160650 33 10067595

962062310 10706260 225655 165640 37 28069420

15 70064350(*) 10 56065475 17456510

15 70063920 13 28064015 17806425

18 17065500 16 09066125 20456615

Activities are expressed as mU / l of plasma. The data are expressed as mean6SD. * P , 0.05; ** P . 0.02; *** P . 0.001: BPD preterm infants before treatment versus reference preterm infants.

G. Goi et al. / Clinica Chimica Acta 278 (1998) 23 – 34

Table 1 Activities of some lysosomal plasma glycohydrolases in bronchopulmonary dysplasia (BPD) preterm infants before and during dexamethasone treatment

G. Goi et al. / Clinica Chimica Acta 278 (1998) 23 – 34

29

Table 2 Diagnostic efficiency analysis of some glycohydrolases of lysosomal origin, according to Bayes’s method Enzyme

Discriminator (mU/l of plasma)

Diagnostic sensitivity

Diagnostic specificity

Diagnostic efficiency

PPpos

PPneg

b-D-Glucuronidase b-D-Galactosidase a-D-Galactosidase a-D-Glucosidase N-Acetyl-b-Dglucosaminidase

. 4850 . 455 . 188 . 215 . 17750

0.45 1.00 0.91 0.45 1.00

0.96 0.68 0.80 0.84 0.48

0.41 0.68 0.71 0.29 0.48

0.83 0.58 0.67 0.56 0.46

0.80 1.00 0.95 0.78 1.00

The definitions of terms used is given in Section 2.6.

we analysed the diagnostic efficiency for b-D-galactosidase in subjects positive at the a-D-galactosidase test (Table 3-A). This enabled us to reduce the number of false positives (Table 3-B). Subsequently, by combined analysis of the two enzymes performed in the total preterm population, our results were improved (Table 3-C).

4. Discussion Bronchopulmonary dysplasia (BPD) is a major cause of morbidity and mortality in premature infants [1–5]. With the improved survival of prematurely

Fig. 1. Activities of some lysosomal glycohydrolases in plasma of preterm infants with bronchopulmonary dysplasia (s) and in reference preterm infants (h). The activities are expressed as U / litre of plasma. N-acetyl-b-D-glucosaminidase: NAG; b-D-glucuronidase: GCR; b-D-galactosidase: BGAL; a-D-galactosidase: aGAL; a-D-glucosidase: aGLU.

30

G. Goi et al. / Clinica Chimica Acta 278 (1998) 23 – 34

Table 3 A Evaluation of a-D-galactosidase test in the total population of 36 preterm infants (prevalence of BPD 0.306) Discriminator, 188 mU / l NEG POS 11 BPD 1 10 Diagnostic sensitivity, 0.91 25 not-BPD 20 5 Diagnostic specificity, 0.80 PPneg PPpos 0.95 0.67 B Evaluation of b-D-galactosidase test in the total a-D-galactosidase positive population (prevalence of BPD 0.670) Discriminator, 468 mU / l 10 BPD 0 10 Diagnostic sensitivity, 1.00 5 not-BPD 2 3 Diagnostic specificity, 0.40 PPneg PPpos 1.00 0.77 C Evaluation of a-D-galactosidase and b-D-galactosidase tests in the total population of 36 preterm infants (prevalence of BPD 0.306) aGAL discriminator, 188 mU / l bGAL discriminator, 468 mU / l 11 BPD 1 10 Diagnostic sensitivity, 0.91 25 non-BPD 22 3 Diagnostic specificity, 0.88 PPneg PPpos 0.96 0.77 The definitions of terms used is given in the Section 2.6. BPD: Bronchopulmonary dysplasia.

born very low birth weight infants, the prevalence of this syndrome is rising and it has become the most common form of chronic lung disease in infants and children. It was recently suggested that lung immaturity is an important pathogenic mechanism underlying BPD [5], and there is increasing evidence that inflammation is also an important factor [28]. It is now known that the activity of some lysosomal glycohydrolases varies during the course of fetal development [6–8]. Data obtained both in humans and in animal models indicate a trend to a gradual decrease in the activity of these enzymes from premature to term and older subjects, which is probably paralleled by a decrease in the number of lysosomes

G. Goi et al. / Clinica Chimica Acta 278 (1998) 23 – 34

31

[6,7,29]. Nevertheless, our understanding of the behaviour of lysosomal enzymes during development is limited. More progress has been made in studies investigating the significance of the same enzymes not only in rare syndromes deriving from an inborn lysosomal disease [30], but also from a number of acquired diseases [13], such as Diabetes mellitus [31] and several physiological and pathological conditions such pregnancy, phagocytosis, immune response, inflammation and tumors [13,15,32,33]. The increased levels of these enzymes in body fluids may also derive from cell lysis or be caused by different compounds (hormones, exogenous lysosomotrophic agents) which specifically interfere with the mechanism of fusion of the intracellular organelles carrying lysosomal enzymes with the plasma membrane. We therefore investigated whether some lysosomal enzymes might be involved in BPD and whether they may play a role in the early identification of BPD patients and / or the monitoring of the effect of therapy using, for example, dexamethasone. According to our findings, the lysosomal apparatus or at least some glycohydrolases of lysosomal origin seem involved in BPD. In fact, except for a-D-glucosidase, plasma levels of all the lysosomal enzymes considered were higher in preterm infants with BPD than in those with only respiratory distress who had not developed BPD. Because the lysosomal apparatus is activated during inflammation, the increase found may have been the result of the inflammatory process in BPD patients [28]. As the pathophysiology of BPD is multifactorial, its treatment is multifaceted and the management of BPD infants currently presents several problems. To date, there are no clinical trials that have definitively demonstrated the effectiveness of commonly used therapies. The use of corticosteroids in the management of BPD is a relatively recent development, and in various studies systemic dexamethasone therapy has been associated with improvements in clinical status and pulmonary function in infants at risk of, or with, BPD [23,34,35]. However, little information is available on the cellular mechanism underlying the activity of dexamethasone and on the effects of prolonged therapies. In the BPD preterm infants monitored in our study, dexamethasone therapy prolonged until 42 days did not appear to change the plasma levels of the lysosomal enzymes considered. Data reported in the literature indicate that in foetal or newborn animals dexamethasone, like other steroids, inhibits the phagocytically induced release of lysosomal enzymes [36], probably stabilising the lysosomal membrane and regulating the extracellular levels of these enzymes, thus stimulating the synthesis of mannose receptors, which play an important role in regulating extracellular levels of lysosomal enzymes via a secretion–recapture mechanism [12]. We found no reduction in the plasma levels of these enzymes in our subjects. This may have depended either on the drug

32

G. Goi et al. / Clinica Chimica Acta 278 (1998) 23 – 34

doses utilised and / or the times at which treatment was given, thus being a potentially important parameter when establishing the modality for treatment administration. Considering that, unlike the forms initially described, BPD now has an insidious onset, with the gradual appearance of persistent abnormalities visualised at X-ray after the first month of life, the availability of a test would be particularly valuable enabling the early identification of subjects at risk who require close observation as this would allow early treatment to be provided where necessary. The data from our retrospective analysis of diagnostic efficiency in the overall population of preterm infants show that at least two of the lysosomal enzymes considered, a-D-galactosidase and b-D-galactosidase, differentiate a posteriori BPD and non-BPD subjects. Further studies focused on preterms at risk for BPD before development of the disease, could provide information about the diagnostic potential of the two above mentioned enzymes. Acknowledgements This work was supported by a grant to Professor Lombardo from the Italian Ministry of Education (40% contribution for 1995–96). References [1] Northway Jr. WH, Rosan RC, Porter DJ. Pulmonary disease following sequelae of bronchopulmonary dysplasia. N EngI J Med 1967;276:357–68. [2] Northway Jr. WH. Bronchopulmonary dysplasia: then and now. Arch Dis Child 1990;65:1076–81. [3] Bancalari E. Pathogenesis of bronchopulmonary dysplasia: an overview. In: Bancalari E, Stocker JT, editors. Bronchopulmonary dysplasia. Washington, DC: Hemisphere, 1988, p. 3–15. [4] Cherukupalli K, Larson JE, Rotschild A, Thurlbeck WM. Biochemical, clinical and morphologic studies on lungs of infants with bronchopulmonary dysplasia. Pediatr Pulmonol 1996;22:215–9. [5] Thurlbeck WM. Prematurity and the developing lung. In: Holtzman RB, Frank L, editors. Clinics in perinatology, vol. 19, Philadelphia: W.B. Saunders Company, 1992, p. 497–520. [6] Heath MF, Jacobson W. Developmental changes in enzyme activities in fetal and neonatal rabbit lung. Cytidyltransferase, cholinephosphotransferase, phospholipase A1 and A2, bgalactosidase and b-glucuronidase. Pediatr Res 1984;18:395–401. [7] Simeoni U, Schinzler B, Massfelder T, et al. Specific developmental profiles of lysosomal and brush border enzymuria in the human. Biol Neonate 1994;65:1–6. [8] Shattuck KE, Richardson CJ, Rassin DK, Lobe TE. Development of serum hexosaminidase activity in infants. Biol of Neonates 1986;49:126–31. [9] Lobe TE, Richardson CJ, Rassin DK, Mills R, Schwartz MZ. Hexosaminidase: a biochemical marker for necrotizing enterocolitis in the preterm infants. Am J Surg 1984;147:49–52. [10] Rajashree S, Puvanakrishnan R. Alterations in certain lysosomal glycohydrolases and cathepsin in rats on dexamethasone administration. Mol cell Biochem 1996;154:165–70.

G. Goi et al. / Clinica Chimica Acta 278 (1998) 23 – 34

33

[11] Hicks JJ, Duran-Reyes G, Diaz-Flores M. Effect of dexamethasone as an inhibitor of implantation and embryo development in rat; lysosomal role. Contraception 1994;50:581–9. [12] Shepherd VL, Konish MG, Stahl P. Dexamethasone increases expression of mannose receptors and decreases extracellular lysosomal enzyme accumulation in macrophages. J Biol Chem 1985;260:160–4. [13] Calvo P, Barba JL, Cabezas JA. Serum b-glucuronidase, b-glucosidase, a-fucosidase, a-galactosidase levels in acute viral hepatitis, pancreatitis, myocardial infarction and breast cancer. Clin Chim Acta 1982;11:15–9. [14] Goi G, Fabi A, Lorenzi R, et al. Serum enzymes of lysosomal origin as indicators of metabolic control in diabetes: comparison with glycated hemoglobin and albumin. Acta Diabetol Lat 1986;23:117–25. [15] Lombardo A, Goi G, Pistolesi E, et al. Behaviour of several enzymes of lysosomal origin in human plasma during pregnancy. Clin Chim Acta 1984;143:253–64. [16] Corral J, Miralles JM, Garcia-Pascual IJ, et al. Increased serum N-acetyl-b-D-glucosaminidase and a-D-mannosidase activities in obese subjects. Clin Invest 1992;70:880–4. [17] Karkkainen P, Poikolainen K, Salaspuro M. Serum b-hexosaminidase as a marker of heavy drinking. Alcohol Clin Exp Res 1990;14:187–90. [18] Goi G, Burlina AB, Moreschi C, et al. Enzymes of lysosomal origin in the serum of infants of diabetic mothers. Behaviour in the first days after birth. Acta Diabetol Lat 1988;25:351– 60. [19] Goi G, Burlina AB, Bairati C, Bordugo A, Zanardo V, Zacchello F. Enzymes of lysosomal origin in plasma of twin neonates. Clin Chim Acta 1993;214:61–71. [20] Lombardo A, Caimi L, Marchesini S, Goi G, Tettamanti G. Enzymes of lysosomal origin in human plasma and serum; assay conditions and parameters influencing the assay. Clin Chim Acta 1980;108:337–46. [21] Goi G, Besozzi M, Bairati C, et al. Assay of enzymes of lysosomal origin in plasma. I. Preparation of a stable liquid material for calibration and quality control for lysosomal enzymes in plasma. Eur J Clin Chem Clin Biochem 1992;30:595–8. [22] Goi G, Bairati C, Roggi C, et al. The lysosomal N-acetyl-b-D-glucosaminidase (NAG) isoenzymes in plasma: study of distribution in a general population by a simple routine chromatofocusing procedure. Clin Chim Acta 1993;221:47–57. [23] Cummings JJ, D’Eugenio DB, Gross SJ. A controlled trial of dexamethasone in preterm infants at high risk from bronchopulmonary dysplasia. N EngI J Med 1989;320:1505–10. [24] Bronchopulmonary dysplasia. In: Guidelines for the care of children with chronic lung disease. Bureau of maternal and child health and resource development. Pediatr. Pulmonol 1989;6(Suppl. 3):3–4. [25] Snedecor GW, Cochram WG. Statistical methods, Ames: Iowa State University Press, 1967. [26] Norusis MJ. SPSS / pc 1 SPSS Inc., Chicago, USA, 1986. [27] Hall GH. The clinical application of Bayes Theorem. The Lancet 1967;Sept:555–557. [28] Watterberg KL, Demers LM, Scott SM, Murphy S. Chorioamnionitis and early lung inflammation in infants in whom bronchopulmonary dyaplasia develops. Pediatrics 1996;97(2):210–5. [29] Oron U. Proteolytic enzyme activity in rat hindlimb muscles in fetus and during post-natal development. Int J Dev Biol 1990;34:457–60. [30] Glew RH, Basu A, Prence EM, Remaly AT. Lysosomal storage diseases. Lab Invest 1985;53:250–69. [31] Burlina AB, Goi G, Fabi A, Lombardo A, Gaburro D, Tettamanti G. Behaviour of some lysosomal enzymes in the plasma of insulin-dependent diabetic patients during artificial pancreas treatment. Clin Biochem 1987;20:423–7. [32] Allen JM. Lysosomes in bacterial infection. In: Dingle JT, Fell HB, editors. Lysosomes in biology and pathology. Amsterdam–London Publishing Company, 1975, p. 41–64.

34

G. Goi et al. / Clinica Chimica Acta 278 (1998) 23 – 34

[33] Henson PM. Secretion of lysosomal enzymes induced by immune complexes and complement. In: Dingle JT, Fell HB, editors. Lysosomes in biology and pathology. Amsterdam– London Publishing Company, 1976, p. 99–126. [34] Rush MG, Hazinski A. Current therapy of bronchopulmonary dysplasia. In: Bancalari E, Stocker JT, editors. Bronchopulmonary dysplasia. Washington, DC: Hemisphere 1988, p. 563–90. [35] Mammel MC, Green TP, Johnson DE, Tompson TR. A controlled trial of dexamethasone therapy in infants with bronchopulmonary dysplasia. Lancet 1983;1(8338):1356–8. [36] Ackerman NR, Beebe JR. Effects of pharmacologic agents on release of lysosomal enzymes from alveolar mononuclear cells. J Pharm Exp Ther 1975;193:603–13.