- Email: [email protected]
Neurodevelopmental outcome after COX inhibitor treatment for patent ductus arteriosus
Early Human Development 86 (2010) 87–92
Contents lists available at ScienceDirect
Early Human Development j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e a r l h u m d ev
Neurodevelopmental outcome after COX inhibitor treatment for patent ductus arteriosus C. Rheinlaender, D. Helfenstein, C. Pees, E. Walch, C. Czernik, M. Obladen, P. Koehne ⁎ Department of Neonatology, Charité, Universitätsmedizin Berlin, Campus Virchow-Klinikum, Germany
a r t i c l e
i n f o
Article history: Received 14 August 2009 Received in revised form 11 December 2009 Accepted 30 December 2009 Keywords: Bayley II scales Griffiths scales Ibuprofen Indomethacin Ligation Neurodevelopmental outcome Patent ductus arteriosus Premature infants Very low birth weight infants
a b s t r a c t Aim: To compare neurodevelopmental results in very low birth weight (VLBW) infants two years after successful or failed cyclooxygenase inhibitor treatment with either indomethacin or ibuprofen for a haemodynamically significant patent ductus arteriosus (hsPDA). Methods: We retrospectively evaluated closure rates and outcome parameters of VLBW infants with hsPDA 89 of whom were treated with indomethacin and 93 with ibuprofen. Results: Indomethacin and ibuprofen therapy groups did not differ in their baseline clinical profile (median gestational age 26.0 and 26.2 wks d) in early (median CRIB 6 and 5, respiratory distress N 2° in 36 and 34 infants) and late morbidities (intraventricular hemorrhage N2° in 9 and 10 infants, bronchopulmonary dysplasia in 31 and 27 infants, 80 and 85 survivors), PDA closure rates (63 and 58%) or neurodevelopmental outcome. The therapy failure group (54 infants) was characterized by lower median gestational age (25.0 wks d) and higher mortality (17%). No differences were found in the neurodevelopmental outcome of the surviving infants with ligation as compared to the survivors with successful pharmacological closure of the PDA at 24 months corrected age. Conclusion: Use of either ibuprofen or indomethacin for closure of a hsPDA did not influence two year neurodevelopmental outcomes in VLBW infants. © 2010 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Treating patent ductus arteriosus (PDA) has been accepted as standard of care in preterm infants to prevent short-term morbidity. The impact of such intervention on infant's neurodevelopmental outcome has, however, recently been questioned [1]. Surgical ligation has been shown to be associated with neonatal morbidities such as bronchopulmonary dysplasia (BPD), retinopathy of prematurity (ROP) and neurosensory impairment in extremely low birth weight (ELBW) infants [2,3]. A meta-analysis of large randomised trials has shown the two COX inhibitors indomethacin and ibuprofen to be equally effective in closing an open duct [4]. Data on neurodevelopmental outcome after prophylactic use of indomethacin are available, but no data on neurodevelopmental follow-up after therapeutic use have been published for either drug [5]. The present comparative historical study was performed to evaluate neurodevelopmental outcome at two years in a cohort of VLBW infants treated with COX inhibitors for a hsPDA. The following two questions were answered. Is the failure of medical PDA closure ⁎ Corresponding author. Department of Neonatology, Charité, University Medical Center, Augustenburger Platz 1, D-13353 Berlin, Germany. Tel.: +49 30 450 566122; fax: +49 30 450 566922. E-mail address: [email protected] (P. Koehne). 0378-3782/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.earlhumdev.2009.12.009
predictive for poor outcome? Does neurological outcome differ between infants treated with ibuprofen and indomethacin? 2. Methods 2.1. Patients In a retrospective study we evaluated closure rates and outcome parameters of 182 VLBW infants (birthweight b 1500 g) undergoing COX inhibitor treatment (89 indomethacin and 93 ibuprofen) for a hsPDA in the Department of Neonatology, Charité Campus VirchowKlinikum between 1998 and 2003 (Fig. 1). Informed parental consent for data collection was obtained after patient admission to our Neonatal Intensive Care Unit. The study has been approved by the local ethics committee. 2.2. PDA intervention In our department COX inhibitor treatment is only initiated in VLBW infants with a hsPDA. The diagnosis is based on strict clinical and echocardiographic criteria. This approach has accounted for a steady PDA intervention rate of 30% over the past decade. While there has been no substantial change in the criteria used to determine the need for PDA intervention since 1998, indomethacin was replaced by ibuprofen in 2001 as an alternative COX inhibitor with fewer
88
C. Rheinlaender et al. / Early Human Development 86 (2010) 87–92
dose regimen of 10–5–5 mg/kg at 24 h intervals, was used to treat PDA [8,9]. 2.6. Patient handling Infants were cared for in closed double wall incubators. Fluid administration was 130 ml/kg/d during PDA intervention. Ventilation was undertaken with a target pCO2 of 35–55 mm Hg and pO2 of 40–60 mm Hg. No substantial changes in patient care have been made during the two treatment periods. 2.7. Neuromotor assessment
Fig. 1. Patient flow after diagnosis of a hsPDA. Numbers are displayed as n (%). Upper boxes display study periods and COX inhibitor therapy groups. Success is defined as successful pharmacological PDA closure. All infants with failed pharmacological PDA closure (with the exception of two from the ibuprofen group who died before ligation) received surgical ligation. Lower boxes display numbers of infants with complete follow-up data at 24 months CA.
undesirable short-term side effects [6,7]. Infants with PDA unresponsive to COX inhibitors received surgical ligation as a rescue therapy. 2.3. PDA diagnosis/echocardiography Infants were examined for a hsPDA at 48 h of life and when clinically indicated. Comprehensive echocardiographic evaluation included shunt direction of a PDA in color Doppler mode (high upper parasternal short axis) as well as the minimal internal ductal diameter of three to five measurements taken in B-mode. The left atrium to aortic root (LA/Ao) ratio was measured by M-mode (parasternal long axis). The Doppler measurement of the resistance index in the anterior cerebral artery was done at the same time. Examination was performed by two experienced echocardiographers using a HDI 3500 (ATL Philips Medical Systems®, Hamburg, Germany) with a curved array transducer of 7.5–10 MHz for twodimensional, M-mode and color Doppler evaluation of the heart and the anterior cerebral artery. 2.4. Algorithm for PDA intervention A PDA with left-to-right shunt was considered haemodynamically significant if (i) a respiratory set back with a supplemental oxygen requirement N 30% and/or mechanical ventilation, (ii) a LA/Ao ratio ≥ 1.4 in the echocardiogram and/or (iii) a decreased enddiastolic flow in the anterior cerebral artery with a resistance index ≥ 0.85 in the cerebral ultrasound was present. Successful response to COX inhibitor treatment was defined as absent ductal shunt flow 24–48 h after therapy, all other cases were defined as COX inhibitor treatment failure. Ligation by clip was performed as a rescue therapy after failure of pharmacological treatment in ventilated infants by two cardio surgeons directly in the NICU in a mean operation time of 20–30 min in the left lateral position. A chest tube was routinely placed to treat postoperative pneumothorax. Infants with successful pharmacological closure of the PDA received routine echocardiographic control before being discharged from the hospital.
Neuromotor assessment included tendon reflexes, muscle tone, early reflexes and locomotor status. EMPP for early hints of cerebral palsy (CP) were evaluated together with a physiotherapist at 6 and 12 months corrected age (CA) [10]. 2.8. Outcome parameters/neurodevelopmental assessment Follow-up examinations were done on infants at two years CA in our outpatient clinic by a neonatologist who specialises in neurodevelopmental testing and a child psychologist. Neurodevelopmental status was evaluated using Griffiths score subscales for motor performance, personal and social behaviour, hearing and speech ability, hand coordination and cognition until March 2005. A test result below 88 points (N2 SD, mean 22 months CA = 98 points, range 96–112 points) indicates severe impairment [11]. From January 2004 neurodevelopmental testing was evaluated with the Bayley Scales II of Infant Development for cognitive, social, speech and gross and fine motor skills. A result of less than 70 points (N2 SD, mean 24 months CA = 100 points, range 85–115 points) indicates severe impairment [12]. A composite poor outcome was defined as the infant's exhibiting one or more of the following at a CA of two years: no free walking, Griffiths DQ score of less than 88, Bayley MDI score of less than 70, bilateral blindness, or bilateral hearing impairment requiring amplification. The following definitions were applied to other outcome parameters: For a diagnosis of BPD we applied the NICHD severity based definition [13,14]. For a ROP diagnosis we applied the ICROP criteria [15]. Intraventricular hemorrhage (IVH) was graded according to Papile [16]. Cystic periventricular leukomalacia (PVL) was defined according to Resch [17]. CP was diagnosed and assessed in keeping with the European consensus paper [18]. Infants underwent cerebral ultrasound examinations on days 1, 3, 7, 14, and 30 of life and before discharge by the attending neonatologist. Auditory evoked potentials were assessed before discharge from hospital and at 6 weeks CA. Unilateral hearing thresholds above 35 dB were considered pathological. Blindness was defined as a corrected visual acuity b20/200. 2.9. Sample size The comparable numbers of infants in both treatment groups resulted from the establishment of ibuprofen as the preferred PDA intervention in the beginning of 2001, which allowed for the analysis of two consecutive time periods prior to and after this change in treatment. 2.10. Statistical models
2.5. COX inhibitor therapy From January 1998 to March 2001 infants received indomethacin, starting with three doses of 0.2 mg/kg in 12 h intervals followed by daily maintenance doses of 0.1 mg/kg for a maximum of 6 days. From April 2001 to December 2003 ibuprofen (Pedea®) with the common
First we examined perinatal and outcome data of the two treatment groups for significant differences. Then we pooled the data from the two therapy groups and examined it for risk factors that led to poor outcome in our patient population using logistic regression.
C. Rheinlaender et al. / Early Human Development 86 (2010) 87–92
In a second approach we identified significant differences in perinatal, PDA-related, neonatal morbidity and follow-up data for the infants with successful pharmacological closure of PDA as compared to infants with failed PDA closure who were all subjected to surgical ligation. Data were analysed using SPSS 14.0. Since the data were not distributed normally, nonparametric statistical tests were employed. Depending on whether the data were continuous, ordinal, or categorical, analysis was carried out using the Mann–Whitney-U test or the chi-square test, respectively. A p-valueb 0.05 was considered statistically significant. 3. Results All together 617 VLBW infants were treated in our unit during the study period. 182 of them developed a PDA that was considered to require intervention. Two year outcome data were available in 141 infants. 3.1. Variables associated with poor outcome in the whole study population In the present study population, composite poor outcome was predicted by several classical risk factors (CRIB score, birth weight, gestational age, gender, number of days with oxygen supplementation, intubation days, ROP, BPD, IVH, PVL, and surfactant administration). In logistic regression the variables IVH N °2 (Rbeta = 0.369, Expbeta = 1.446, p = 0.033), intubation days (Rbeta = 0.046, Expbeta = 1.048, p = 0.004) and male sex (Rbeta =−0.905, Expbeta = 0.405, p = 0.021) were negatively associated with neurodevelopmental outcome. Failure of COX inhibitor therapy (Rbeta = −0.013, Expbeta = 0.987, p = 0.973) and PDA duration (Rbeta = 0.017, Expbeta = 1.017, p = 0.497) had no independent effect on composite poor outcome. 3.2. COX inhibitor therapy success vs. failure COX inhibitor treatment was not successful in 54 infants (30%). These infants except two who died before ligation could be performed, received ligation as a rescue therapy. These infants were significantly more premature and smaller, had worse CRIB scores and were at higher risk for death before 24 months CA. The number of infants with mechanical ventilation and rates of BPD and ROP was higher, the hospitalization period longer (Table 1). No differences were found due to gender and rates of IVH, PVL and NEC. Echocardiographic parameters (maximum PDA diameter, number of infants with PDA diameter N 1.5 mm, left atrium to aortic root ratio N 1.4 and/or resistance index N 0.85) showed no differences before intervention in both groups. The data revealed major differences between the infants with successful COX inhibitor therapy and those where therapy failed with respect to early follow-up data (Table 1). No differences were found in neurodevelopmental test results at 24 months CA (Table 2). In the therapy failure group the follow-up recall rate is significantly lower than in the success group, 78% vs. 92% respectively. Reasons for missing data were 1) removal/relocation, 2) tests performed in another outpatient clinic in Berlin, 3) home address outside Berlin and 4) incompliance. The percentage of the later two were higher in the treatment failure group (each 7% = 3 infants) than in the treatment success group (each 3% = 3 infants). To rule out a systematic effect due to the change in the use of the Griffiths to the Bayley scales during the study period, we analysed the data subsets of each test separately. In the Griffiths DQ 79 infants of the treatment success group (median 92 points; Q 1/3: 83/100) and 23 infants of the treatment failure group (median 92 points; Q 1/3: 89/96, p N 0.16) were tested at 22 months. In the Bayley test 32 infants of the treatment success group (median 90 points; Q 1/3: 75/99) and 13
89
Table 1 Socioeconomic characteristics of mothers, perinatal, PDA treatment and neonatal morbidity data summarized according to therapy success and failure. COX inhibitor therapy Success (128 infants) Socioeconomic data Age Level of education, n (%) No school certificate Nine school years Ten–twelve school years University qualification Unknown Single parent, n (%) Perinatal data Birthweight (g) GA (wks d) CRIB score Female, n (%) Completed, n (%) Prenatal steroids, n (%) Cesarean delivery, n (%) Cord blood pH RDS N 2°, n (%) Surfactant, n (%) Mech.ventilation, n (%) Ventilator days (ventilated infants only) FiO2 ≥ 0.30, n (%)
Neonatal morbidity NEC, n (%) ROP N 2°, n (%) Suppl. O2 wk. 36, n (%) IVH N 2°, n (%) PVL, n (%) Hospitalization days Survivors, n (%)
29 (24.5–33.5)
27 (23–32)
25 18 54 30 1 17
4 (7) 10 (19) 23 (43) 15 (28) 2 (4) 7 (13)
883 26.4 5 52 89 107 96 7.3 40 88 65 8
PDA treatment data First intervention (dol) Exposition to PDA (d) Reopening by infection, n (%)
Failure (54 infants)
(19) (14) (42) (23) (1) (13)
(770–1149) (25.5–28.1) (2–7) (41) (70) (84) (75) (7.25–7.34) (32) (69) (51) (4–18)
750 (665–881) 25.0 (24.3–26.5) 7 (4–10) 24 (44) 35 (65) 45 (83) 42 (78) 7.3 (7.24–7.35) 30 (58) 47 (87) 41 (76) 18 (12–26)
p
0.06 0.16
0.95
0.001 0.001 0.001 0.6 0.5 0.6 0.6 0.3 0.001 0.01 0.003 0.001
41 (32)
23 (43)
0.2
4 (3–6) 4 (3–5) 15 (12)
4 (2–7) 8 (6–11) 18 (33)
0.9 0.001 0.001
5 (10) 9 (17) 31 (57) 5 (10) 6 (11) 101 (82–131) 45 (83)
0.95 0.001 0.001 0.48 0.18 0.001 0.03
16 5 27 14 7 78 120
(13) (4) (21) (11) (6) (61–96) (94)
Median and quartiles are displayed, if not otherwise indicated. GA = gestational age, dol = day of life, RDS = respiratory distress syndrome, IVH = intraventricular hemorrhage, NEC = necrotizing enterocolitis, ROP = retinopathy of prematurity, PVL = cystic periventricular leukomalacia, and survivors = discharged home after primary hospital stay.
infants of the treatment failure group (median 80 points; Q 1/3: 49/87, p N 0.06) were tested at 24 months. No differences were found in hearing and walking ability or composite poor outcome at 24 months CA (Table 2).
Table 2 Follow-up data summarized according to therapy success and failure. Followed at two years
Hearing level N 35 dB, n (%) Hearing aid Blind Free walking at 2 years, n (%) CP Griffiths DQ 22 months b 88 or Bayley MDI 24 months b 70, n (%) Composite poor outcome
COX inhibitor therapy Success (109 infants)
Failure (32 infants)
p
11(10) 1 (1) 1 (1) 103 (95) 9 (5) 31 (29)
3 (9) 1 (1) 1 (1) 30 (94) 5 (16) 7 (22)
0.8 – – 0.70 0.24 0.50
37 (34)
8 (25)
0.40
Only data of infants with complete follow-up data are shown.
90
C. Rheinlaender et al. / Early Human Development 86 (2010) 87–92
3.3. Ibuprofen vs. indomethacin
Table 4 Follow-up data of both COX inhibitor therapy groups.
Comparison of infants who were treated with ibuprofen or indomethacin revealed no differences in their baseline clinical profile apart from a higher number of female individuals in the indomethacin group. No differences were found in perinatal and early and late morbidity data (Table 3). The number of infants with therapy related oliguria (indomethacin 3 and ibuprofen 1) and thrombocytopenia (indomethacin 16 and ibuprofen 6) was small in both groups. PDA closure rates and neurodevelopmental outcome at two years were alike in both groups (Tables 3 and 4). 4. Discussion
Table 3 Socioeconomic characteristics of mothers, perinatal, PDA treatment and neonatal morbidity data of both COX inhibitor therapy groups. COX inhibitor therapy Indomethacin (89 patients)
Perinatal data Birthweight (g) GA (wks d) CRIB score Female, n (%) Prenatal steroids, n (%) Completed, n (%) Cesarean delivery, n (%) Cord blood pH RDS N 2°, n (%) Surfactant, n (%) Mech.ventilation, n (%) Ventilator days (ventilated infants only) FiO2 ≥ 0.30, n (%)
Hearing level N 35 dB, n (%) Hearing aid Blind Free walking at 24 months, n (%) CP Griffiths DQ 22 months b 88 or Bayley MDI 24 months b 70, n (%) Composite poor outcome
COX inhibitor therapy Indomethacin (71 infants)
Ibuprofen (70 infants)
p
7 (10) 0 (0) 1 (1) 67 (94) 8 (11) 23 (32)
7 2 1 67 6 15
(10) (3) (1) (96) (9) (22)
0.81 – – 0.71 0.65 0.14
27 (38)
19 (28)
0.17
Only data of infants with complete follow-up data are shown.
The question of the impact of PDA ligation in preterm infants has been thoroughly discussed recently, as has the issue of the preferred drug. In an earlier study on the short-term complications of indomethacin treatment and PDA ligation, we proposed that surgical ligation should be reserved for infants not responding to pharmacological PDA closure [19]. With our current study, we aimed to add further information to this discussion by a detailed data analysis of our
Socioeconomic data Age Level of education, n (%) No school certificate Nine school years Ten–twelve school years University qualification Unknown Single parent, n (%)
Followed at two years
Ibuprofen (93 patients)
27 (24–33)
29 (24–33)
12 (14) 19 (21) 35 (39) 22 (25) 1 (1) 10 (11)
17 9 42 23 2 14
845 (730–1072) 26.0 (25.0–28.0) 6 (2–9) 44 (49) 70 (79) 62 (70) 68 (76) 7.3 (7.26–7.34) 36 (40) 67 (75) 54 (62) 15 (7–21)
850 26.2 5 32 82 62 70 7.3 34 68 52 15
p
0.42 0.23
(18) (10) (45) (25) (2) (15)
0.45
(717–1127) (25.0–27.6) (3–8) (34) (88) (67) (75) (7.23–7.34) (37) (73) (57) (4–11)
0.70 0.54 0.35 0.04 0.03 0.57 0.96 0.66 0.19 0.74 0.17 0.17
33 (37)
31 (33)
0.60
PDA treatment data First intervention (dol) Exposition to PDA (d) Reopening by infection, n (%) Treatment failure, n (%)
4 (3–7) 4 (4–6) 17 (19) 24 (27)
3 4 12 30
(2–6) (3–9) (13) (32)
0.65 0.47 0.24 0.44
Neonatal morbidity NEC, n (%) ROP N 2°, n (%) Suppl. O2 wk. 36, n (%) IVH N 2°, n (%) PVL, n (%) Hospitalization days Survivors, n (%)
10 (11) 9 (10) 31 (35) 10 (11) 7 (8) 89 (72–106) 80 (90)
11 5 27 9 6 76 85
(12) (5) (29) (10) (7) (60–96) (91)
0.20 0.21 0.39 0.47 0.71 0.07 0.91
Median and quartiles are displayed, if not otherwise indicated. GA = gestational age, dol = day of life, RDS = respiratory distress syndrome, IVH = intraventricular hemorrhage, NEC = necrotizing enterocolitis, ROP = retinopathy of prematurity, PVL = cystic periventricular leukomalacia, and survivors = discharged home after primary hospital stay.
hsPDA patient population. The flaws and disadvantages of the historical design are partly counterbalanced by strict therapy standards and examining staff that remained largely unchanged during the study period as well as a low rate of infants lost to followup. They are also mitigated by the fact that, with the exception of a change in preferred drug treatment, general management strategies for PDA did not undergo significant changes at our clinic in the specified time period, leading us to conclude that their effect on our data is inconsequential. The change from Griffiths DQ to Bayley scales for neurodevelopmental testing in accordance with international standards during the study period is one limitation of our study. Since the numerical scores of the two test sets are not equivalent, it is not possible to compare median test results directly. No comparative study of both tests has been done for preterm infants. It has, however, been demonstrated in other patient populations that both tests can be used interchangeably [20,21]. In the period where both test methods were used simultaneously, the outcome of 11 infants was equally categorized as either normal (7 infants) or poor (4 infants). The separate analysis of the median Griffiths and Bayley test results also revealed no statistical differences, though the significance of this is limited due to the small number of infants tested. The use of the same examiners over the whole study period is an advantage that allows us to consider our follow-up analysis reliable and comparable despite the use of two different tests. In our study, failed PDA closure following COX inhibitor therapy was associated with prematurity, neonatal morbidity and mortality, but not with poor neurodevelopmental outcome of the surviving infants at two years CA. Both the Schmidt and Clyman groups have published prospective studies on that issue [2,3]. The perinatal data of their and our study have consistently shown that the smallest, most premature infants are those that require surgical ligation after failed PDA closure by COX inhibitor therapy. In a retrospective study Noori's group has shown that the failure of ductal closure is associated with an increase in mortality in very preterm infants [22]. Failed PDA closure is associated with morbidity represented by prolonged hospitalization and higher mortality in our study population as well. Not surprisingly more infants in the therapy failure group were mechanically ventilated, given that respiratory setback (reintubation, FiO2 N 30%) was one criteria for the diagnosis of a hsPDA. Our echocardiographic data showed equal haemodynamic stress by PDA before intervention. PDA duration was longer in the therapy failure group supposedly because of the time needed to arrange for surgical ligation. The higher rate of reopening due to infection is probably caused by the immaturity and hence relative unresponsiveness of the PDA to COX inhibitors and its additional exposition to higher COX activity and prostaglandins. An efficacy of 60% closure rate after COX inhibitor therapy leaves room for improvement, as a higher success rate would probably lower ligation rate and hence could lead to better outcome results.
C. Rheinlaender et al. / Early Human Development 86 (2010) 87–92
Prolongation of COX inhibitor therapy as well as the escalation of doses has been discussed and investigated recently [23–25]. In our study a prolonged indomethacin protocol was used with 0.1 mg/kg up to six days but did not improve the rate of successful PDA closure in comparison to the ibuprofen group treated with the standard dosing of 10–5–5 mg/kg at 24 h intervals. Sperandio's group followed an interesting protocol of indomethacin dose titration up to 1 mg/kg in nonresponders after a conventional 3-dose course with 0.2 mg/kg indomethacin every 12 h until ductal closure was achieved in 98% of the infants without severe side effects [23]. Unfortunately, this retrospective analysis provided no information on the ROP rate. Clyman's group detected an association between the incidence of moderate and severe ROP and high indomethacin doses of 0.2 or 0.5 mg/kg for an extended 3-day course without a significant decrease in the rate of PDA in a multicenter randomised controlled trial, no recommendation was given on higher indomethacin doses for the prolonged treatment of PDA [24]. Overmeire's group investigated the association between areas under the curve for three conventional ibuprofen doses and successful PDA closure in VLBW infants and suggested a dosing scheme up to 18, 9, 9 mg/kg depending on postnatal age [25]. However, these theoretical findings still need to be tested prospectively. The evident differences of clinical data of the therapy failure compared to the success group lead to the expectation that neurodevelopmental outcome might differ as well, but surprisingly, we saw no statistical difference in neurodevelopmental outcome at 24 months CA. This result was not due to death of the most severely impaired infants during their second year of life. All together 5 infants died during the follow-up period, one in the therapy success group (persistent pulmonary hypertension) and 4 infants in the therapy failure group (1 diagnosis unknown, 2 infants with unknown syndrome, and 1 SIDS). The inclusion of death as a variable to composite poor outcome did not result in a significant difference. We can only presume that some infants caught up in their developmental delays. Furthermore, birth weight and CRIB score have a higher predictive power for neonatal morbidity and mortality than for neurodevelopmental outcome [26]. In the therapy failure group the follow-up recall rate is significantly lower than in the success group, 78% vs. 92% respectively. The fact that poor outcome is associated with parental incompliance for different reasons is nothing new — we should therefore not rule out the possibility that a higher follow-up rate would change outcome results, as the Clyman work has shown [3]. Additionally an extended follow-up assessment at 5 years corrected age will contribute to this question and might identify significant differences which are not apparent at two years. It seems almost impossible to separate the effect of surgical ligation from that of perinatal morbidity on the outcome of preterm infants. The group of Schmidt found that surgical PDA ligation in ELBW infants is associated with BPD, ROP and neurosensory impairment and suspected it of being a cause of a poor outcome at 18 months CA [2]. However, the meta-analysis of Fowlie and Davis did not show improved neurodevelopmental outcome despite a reduction in the rate of ligation in ELBW infants with prophylactic indomethacin therapy [1]. We also found a significantly higher rate of BPD and ROP in infants who underwent ligation after failed pharmacological PDA intervention. The Clyman group with a study population twice as large as ours demonstrated no effect of surgical ligation on neurodevelopmental impairment, but showed an independent effect on chronic lung disease [3]. We agree with the Brooks group that a randomised controlled trial of surgical PDA ligation after failed pharmacological treatment is required to clarify the role of this intervention on neurodevelopmental outcome in VLBW infants [27]. Until now no randomised controlled trial has provided data on neurodevelopmental follow-up at two years after therapeutic use for either COX inhibitor. In our study neurodevelopmental outcome did not differ after the use of either indomethacin or ibuprofen for
91
pharmacological closure of a hsPDA in VLBW infants. Several comparative studies account for equal efficacy of indomethacin and ibuprofen in closing a PDA, but the two COX inhibitors differ slightly in their side effects [4,28]. Indomethacin is known to cause higher decreases in renal and cerebral blood flow, and ibuprofen use has been associated with pulmonary hypertension, higher total bilirubin concentration and CLD [7,28,29]. At any rate, these different pharmacological profiles and the varied undesirable side effects associated with each drug, the major reason neonatologists switched to ibuprofen some years ago, did not seem to have relevant impact on neurodevelopmental outcome in our study population. Since we detected no significant differences in neurodevelopmental outcome at two years between the two COX inhibitor therapy groups, this retrospective data analysis does not indicate whether indomethacin or ibuprofen is preferable for the pharmacological therapy of a PDA. 5. Conclusion For the first time we were able to demonstrate that treatment for a hsPDA in preterm infants with either indomethacin or ibuprofen does not result in a difference in mortality or severely impaired neurodevelopmental outcome assessed at 2 years CA. Immaturity and mortality are associated with COX inhibitor treatment failure and hence ligation. Acknowledgements This study was supported by Orphan Europe and a Rahel-Hirschresearch grant (Dr. Koehne). We thank Boris Metze for the technical and statistical support. We thank Sarah Smithson-Compton for the suggestions regarding the English language. None of the authors has to declare a conflict of interest. References [1] Fowlie PW, Davis PG. Prophylactic intravenous indomethacin for preventing mortality and morbidity in preterm infants. Cochrane Database Syst Rev 2002;3: CD000174. [2] Kabra NS, Schmidt B, Roberts RS, Doyle LW, Papile L, Fanaroff A, et al. Neurosensory impairment after surgical closure of patent ductus arteriosus in extremely low birth weight infants: results from the Trial of Indomethacin Prophylaxis in Preterms. J Pediatr 2007;150:229–34. [3] Chorne N, Leonard C, Piecuch R, Clyman RI. Patent ductus arteriosus and its treatment as risk factors for neonatal and neurodevelopmental morbidity. Pediatrics 2007;119:1165–74. [4] Van Overmeire B, Chemtob S. The pharmacologic closure of the patent ductus arteriosus. Semin Fetal Neonatal Med 2005;10:177–84. [5] Couser RJ, Hoekstra RE, Ferrara TB, Wright GB, Cabalka AK, Connett JE. Neurodevelopmental follow-up at 36 months' corrected age of preterm infants treated with prophylactic indomethacin. Arch Pediatr Adolesc Med 2000;154(6): 598–602. [6] Aranda JV, Thomas R. Systematic review: intravenous Ibuprofen in preterm newborns. Semin Perinatol 2006;30:114–20. [7] Thomas RL, Parker GC, Van Overmeire B, Aranda JV. A meta-analysis of ibuprofen versus indomethacin for closure of patent ductus arteriosus. Eur J Pediatr 2005;164:135–40. [8] Safety Survey Report Named Patient Use of Ibuprofen Orphan Europe (PEDEA) Raport IBU/SURVEY/April 2003. [9] Lago P, Bettiol T, Salvadori S, Pitassi I, Vianello A, Chiandetti L, et al. Safety and efficacy of ibuprofen versus indomethacin in preterm infants treated for patent ductus arteriosus: a randomised controlled trial. Eur J Pediatr 2002;161:202–7. [10] Morgan AM, Aldag JC. Early identification of cerebral palsy using a profile of abnormal motor patterns. Pediatrics 1996;98(4 Pt 1):692–7. [11] Brandt I, Sticker E. Griffith Entwicklungsskalen (GES) zur Beurteilung der Entwicklung in den ersten beiden Lebensjahren. 2. erweiterte und überarbeitete Version. Göttingen: Beltz Test-GmbH; 2001. [12] Bayley N. Bayley scales of infant development (Bsid-II). 3rd ed. Sydney: Pearson Psychcorp; 1999. [13] Jobe AH, Bancalari E. Bronchopulmonary dysplasia. Am J Respir Crit Care Med 2001;163:1723–9. [14] Ehrenkranz RA, Walsh MC, Vohr BR, Jobe AH, Wright LL, Fanaroff AA, et al. Validation of the National Institutes of Health consensus definition of bronchopulmonary dysplasia. Pediatrics 2005;116:1353–60. [15] The International Committee for the Classification of the Late Stages of Retinopathy of Prematurity. An international classification of retinopathy of
92
[16]
[17]
[18]
[19]
[20] [21]
[22]
[23]
C. Rheinlaender et al. / Early Human Development 86 (2010) 87–92 prematurity. II. The classification of retinal detachment. Arch Ophthalmol 1987;105:906–12. Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr 1978;92:529–34. Resch B, Vollaard E, Maurer U, Haas J, Rosegger H, Müller W. Risk factors and determinants of neurodevelopmental outcome in cystic periventricular leucomalacia. Eur J Pediatr 2000;159:663–70. Surveillance of cerebral palsy in Europe: a collaboration of cerebral palsy surveys and registers. Surveillance of Cerebral Palsy in Europe (SCPE). Dev Med Child Neurol 2000;42(12):816–24. Koehne PS, Bein G, Alexi-Meskhishvili V, Wenig Y, Bührer C, Obladen M. Patent ductus arteriosus in very low birthweight infants: complications of pharmacological and surgical treatment. J Perinat Med 2001;29:327–34. Ramsay M, Fitzhardinge PM. A comparative study of two developmental scales: the Bayley and the Griffiths. Early Hum Dev 1977;1(2):151–7. Beail N. A comparative study of profoundly multiply handicapped children's scores on the Bayleys and Griffiths developmental scales. Child Care Health Dev 1985;11:31–6. Noori S, McCoy M, Friedlich P, Bright B, Gottipati V, Seri I, et al. Failure of ductus arteriosus closure is associated with increased mortality in preterm infants. Pediatrics 2009;123:e138–44. Sperandio M, Beedgen B, Feneberg R, Huppertz C, Brüssau J, Pöschl J, et al. Effectiveness and side effects of an escalating, stepwise approach to indomethacin
[24]
[25]
[26]
[27]
[28]
[29]
treatment for symptomatic patent ductus arteriosus in premature infants below 33 weeks of gestation. Pediatrics 2005;116:1361–6. Jegatheesan P, Ianus V, Buchh B, Yoon G, Chorne N, Ewig A, et al. Increased indomethacin dosing for persistent patent ductus arteriosus in preterm infants: a multicenter, randomized, controlled trial. J Pediatr 2008;153:183–9. Hirt D, Van Overmeire B, Treluyer JM, Langhendries JP, Marguglio A, Eisinger MJ, et al. An optimized ibuprofen dosing scheme for preterm neonates with patent ductus arteriosus, based on a population pharmacokinetic and pharmacodynamic study. Br J Clin Pharmacol 2008;65:629–36. Buhrer C, Grimmer I, Metze B, Obladen M. The CRIB (Clinical Risk Index for Babies) score and neurodevelopmental impairment at one year corrected age in very low birth weight infants. Intensive Care Med 2000;26(3):325–9. Brooks JM, Travadi JN, Patole SK, Doherty DA, Simmer K. Is surgical ligation of patent ductus arteriosus necessary? The Western Australian experience of conservative management. Arch Dis Child Fetal Neonatal Ed 2005;90:F235–9. Ohlsson A, Walia R, Shah S. Ibuprofen for the treatment of patent ductus arteriosus in preterm and/or low birth weight infants. Cochrane Database Syst Rev 2008;1: CD003481. Rheinlaender C, Helfenstein D, Walch E, Berns M, Obladen M, Koehne P. Total serum bilirubin levels during cyclooxygenase inhibitor treatment for patent ductus arteriosus in preterm infants. Acta Paediatr 2009;98:36–42.
Contents lists available at ScienceDirect
Early Human Development j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e a r l h u m d ev
Neurodevelopmental outcome after COX inhibitor treatment for patent ductus arteriosus C. Rheinlaender, D. Helfenstein, C. Pees, E. Walch, C. Czernik, M. Obladen, P. Koehne ⁎ Department of Neonatology, Charité, Universitätsmedizin Berlin, Campus Virchow-Klinikum, Germany
a r t i c l e
i n f o
Article history: Received 14 August 2009 Received in revised form 11 December 2009 Accepted 30 December 2009 Keywords: Bayley II scales Griffiths scales Ibuprofen Indomethacin Ligation Neurodevelopmental outcome Patent ductus arteriosus Premature infants Very low birth weight infants
a b s t r a c t Aim: To compare neurodevelopmental results in very low birth weight (VLBW) infants two years after successful or failed cyclooxygenase inhibitor treatment with either indomethacin or ibuprofen for a haemodynamically significant patent ductus arteriosus (hsPDA). Methods: We retrospectively evaluated closure rates and outcome parameters of VLBW infants with hsPDA 89 of whom were treated with indomethacin and 93 with ibuprofen. Results: Indomethacin and ibuprofen therapy groups did not differ in their baseline clinical profile (median gestational age 26.0 and 26.2 wks d) in early (median CRIB 6 and 5, respiratory distress N 2° in 36 and 34 infants) and late morbidities (intraventricular hemorrhage N2° in 9 and 10 infants, bronchopulmonary dysplasia in 31 and 27 infants, 80 and 85 survivors), PDA closure rates (63 and 58%) or neurodevelopmental outcome. The therapy failure group (54 infants) was characterized by lower median gestational age (25.0 wks d) and higher mortality (17%). No differences were found in the neurodevelopmental outcome of the surviving infants with ligation as compared to the survivors with successful pharmacological closure of the PDA at 24 months corrected age. Conclusion: Use of either ibuprofen or indomethacin for closure of a hsPDA did not influence two year neurodevelopmental outcomes in VLBW infants. © 2010 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Treating patent ductus arteriosus (PDA) has been accepted as standard of care in preterm infants to prevent short-term morbidity. The impact of such intervention on infant's neurodevelopmental outcome has, however, recently been questioned [1]. Surgical ligation has been shown to be associated with neonatal morbidities such as bronchopulmonary dysplasia (BPD), retinopathy of prematurity (ROP) and neurosensory impairment in extremely low birth weight (ELBW) infants [2,3]. A meta-analysis of large randomised trials has shown the two COX inhibitors indomethacin and ibuprofen to be equally effective in closing an open duct [4]. Data on neurodevelopmental outcome after prophylactic use of indomethacin are available, but no data on neurodevelopmental follow-up after therapeutic use have been published for either drug [5]. The present comparative historical study was performed to evaluate neurodevelopmental outcome at two years in a cohort of VLBW infants treated with COX inhibitors for a hsPDA. The following two questions were answered. Is the failure of medical PDA closure ⁎ Corresponding author. Department of Neonatology, Charité, University Medical Center, Augustenburger Platz 1, D-13353 Berlin, Germany. Tel.: +49 30 450 566122; fax: +49 30 450 566922. E-mail address: [email protected] (P. Koehne). 0378-3782/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.earlhumdev.2009.12.009
predictive for poor outcome? Does neurological outcome differ between infants treated with ibuprofen and indomethacin? 2. Methods 2.1. Patients In a retrospective study we evaluated closure rates and outcome parameters of 182 VLBW infants (birthweight b 1500 g) undergoing COX inhibitor treatment (89 indomethacin and 93 ibuprofen) for a hsPDA in the Department of Neonatology, Charité Campus VirchowKlinikum between 1998 and 2003 (Fig. 1). Informed parental consent for data collection was obtained after patient admission to our Neonatal Intensive Care Unit. The study has been approved by the local ethics committee. 2.2. PDA intervention In our department COX inhibitor treatment is only initiated in VLBW infants with a hsPDA. The diagnosis is based on strict clinical and echocardiographic criteria. This approach has accounted for a steady PDA intervention rate of 30% over the past decade. While there has been no substantial change in the criteria used to determine the need for PDA intervention since 1998, indomethacin was replaced by ibuprofen in 2001 as an alternative COX inhibitor with fewer
88
C. Rheinlaender et al. / Early Human Development 86 (2010) 87–92
dose regimen of 10–5–5 mg/kg at 24 h intervals, was used to treat PDA [8,9]. 2.6. Patient handling Infants were cared for in closed double wall incubators. Fluid administration was 130 ml/kg/d during PDA intervention. Ventilation was undertaken with a target pCO2 of 35–55 mm Hg and pO2 of 40–60 mm Hg. No substantial changes in patient care have been made during the two treatment periods. 2.7. Neuromotor assessment
Fig. 1. Patient flow after diagnosis of a hsPDA. Numbers are displayed as n (%). Upper boxes display study periods and COX inhibitor therapy groups. Success is defined as successful pharmacological PDA closure. All infants with failed pharmacological PDA closure (with the exception of two from the ibuprofen group who died before ligation) received surgical ligation. Lower boxes display numbers of infants with complete follow-up data at 24 months CA.
undesirable short-term side effects [6,7]. Infants with PDA unresponsive to COX inhibitors received surgical ligation as a rescue therapy. 2.3. PDA diagnosis/echocardiography Infants were examined for a hsPDA at 48 h of life and when clinically indicated. Comprehensive echocardiographic evaluation included shunt direction of a PDA in color Doppler mode (high upper parasternal short axis) as well as the minimal internal ductal diameter of three to five measurements taken in B-mode. The left atrium to aortic root (LA/Ao) ratio was measured by M-mode (parasternal long axis). The Doppler measurement of the resistance index in the anterior cerebral artery was done at the same time. Examination was performed by two experienced echocardiographers using a HDI 3500 (ATL Philips Medical Systems®, Hamburg, Germany) with a curved array transducer of 7.5–10 MHz for twodimensional, M-mode and color Doppler evaluation of the heart and the anterior cerebral artery. 2.4. Algorithm for PDA intervention A PDA with left-to-right shunt was considered haemodynamically significant if (i) a respiratory set back with a supplemental oxygen requirement N 30% and/or mechanical ventilation, (ii) a LA/Ao ratio ≥ 1.4 in the echocardiogram and/or (iii) a decreased enddiastolic flow in the anterior cerebral artery with a resistance index ≥ 0.85 in the cerebral ultrasound was present. Successful response to COX inhibitor treatment was defined as absent ductal shunt flow 24–48 h after therapy, all other cases were defined as COX inhibitor treatment failure. Ligation by clip was performed as a rescue therapy after failure of pharmacological treatment in ventilated infants by two cardio surgeons directly in the NICU in a mean operation time of 20–30 min in the left lateral position. A chest tube was routinely placed to treat postoperative pneumothorax. Infants with successful pharmacological closure of the PDA received routine echocardiographic control before being discharged from the hospital.
Neuromotor assessment included tendon reflexes, muscle tone, early reflexes and locomotor status. EMPP for early hints of cerebral palsy (CP) were evaluated together with a physiotherapist at 6 and 12 months corrected age (CA) [10]. 2.8. Outcome parameters/neurodevelopmental assessment Follow-up examinations were done on infants at two years CA in our outpatient clinic by a neonatologist who specialises in neurodevelopmental testing and a child psychologist. Neurodevelopmental status was evaluated using Griffiths score subscales for motor performance, personal and social behaviour, hearing and speech ability, hand coordination and cognition until March 2005. A test result below 88 points (N2 SD, mean 22 months CA = 98 points, range 96–112 points) indicates severe impairment [11]. From January 2004 neurodevelopmental testing was evaluated with the Bayley Scales II of Infant Development for cognitive, social, speech and gross and fine motor skills. A result of less than 70 points (N2 SD, mean 24 months CA = 100 points, range 85–115 points) indicates severe impairment [12]. A composite poor outcome was defined as the infant's exhibiting one or more of the following at a CA of two years: no free walking, Griffiths DQ score of less than 88, Bayley MDI score of less than 70, bilateral blindness, or bilateral hearing impairment requiring amplification. The following definitions were applied to other outcome parameters: For a diagnosis of BPD we applied the NICHD severity based definition [13,14]. For a ROP diagnosis we applied the ICROP criteria [15]. Intraventricular hemorrhage (IVH) was graded according to Papile [16]. Cystic periventricular leukomalacia (PVL) was defined according to Resch [17]. CP was diagnosed and assessed in keeping with the European consensus paper [18]. Infants underwent cerebral ultrasound examinations on days 1, 3, 7, 14, and 30 of life and before discharge by the attending neonatologist. Auditory evoked potentials were assessed before discharge from hospital and at 6 weeks CA. Unilateral hearing thresholds above 35 dB were considered pathological. Blindness was defined as a corrected visual acuity b20/200. 2.9. Sample size The comparable numbers of infants in both treatment groups resulted from the establishment of ibuprofen as the preferred PDA intervention in the beginning of 2001, which allowed for the analysis of two consecutive time periods prior to and after this change in treatment. 2.10. Statistical models
2.5. COX inhibitor therapy From January 1998 to March 2001 infants received indomethacin, starting with three doses of 0.2 mg/kg in 12 h intervals followed by daily maintenance doses of 0.1 mg/kg for a maximum of 6 days. From April 2001 to December 2003 ibuprofen (Pedea®) with the common
First we examined perinatal and outcome data of the two treatment groups for significant differences. Then we pooled the data from the two therapy groups and examined it for risk factors that led to poor outcome in our patient population using logistic regression.
C. Rheinlaender et al. / Early Human Development 86 (2010) 87–92
In a second approach we identified significant differences in perinatal, PDA-related, neonatal morbidity and follow-up data for the infants with successful pharmacological closure of PDA as compared to infants with failed PDA closure who were all subjected to surgical ligation. Data were analysed using SPSS 14.0. Since the data were not distributed normally, nonparametric statistical tests were employed. Depending on whether the data were continuous, ordinal, or categorical, analysis was carried out using the Mann–Whitney-U test or the chi-square test, respectively. A p-valueb 0.05 was considered statistically significant. 3. Results All together 617 VLBW infants were treated in our unit during the study period. 182 of them developed a PDA that was considered to require intervention. Two year outcome data were available in 141 infants. 3.1. Variables associated with poor outcome in the whole study population In the present study population, composite poor outcome was predicted by several classical risk factors (CRIB score, birth weight, gestational age, gender, number of days with oxygen supplementation, intubation days, ROP, BPD, IVH, PVL, and surfactant administration). In logistic regression the variables IVH N °2 (Rbeta = 0.369, Expbeta = 1.446, p = 0.033), intubation days (Rbeta = 0.046, Expbeta = 1.048, p = 0.004) and male sex (Rbeta =−0.905, Expbeta = 0.405, p = 0.021) were negatively associated with neurodevelopmental outcome. Failure of COX inhibitor therapy (Rbeta = −0.013, Expbeta = 0.987, p = 0.973) and PDA duration (Rbeta = 0.017, Expbeta = 1.017, p = 0.497) had no independent effect on composite poor outcome. 3.2. COX inhibitor therapy success vs. failure COX inhibitor treatment was not successful in 54 infants (30%). These infants except two who died before ligation could be performed, received ligation as a rescue therapy. These infants were significantly more premature and smaller, had worse CRIB scores and were at higher risk for death before 24 months CA. The number of infants with mechanical ventilation and rates of BPD and ROP was higher, the hospitalization period longer (Table 1). No differences were found due to gender and rates of IVH, PVL and NEC. Echocardiographic parameters (maximum PDA diameter, number of infants with PDA diameter N 1.5 mm, left atrium to aortic root ratio N 1.4 and/or resistance index N 0.85) showed no differences before intervention in both groups. The data revealed major differences between the infants with successful COX inhibitor therapy and those where therapy failed with respect to early follow-up data (Table 1). No differences were found in neurodevelopmental test results at 24 months CA (Table 2). In the therapy failure group the follow-up recall rate is significantly lower than in the success group, 78% vs. 92% respectively. Reasons for missing data were 1) removal/relocation, 2) tests performed in another outpatient clinic in Berlin, 3) home address outside Berlin and 4) incompliance. The percentage of the later two were higher in the treatment failure group (each 7% = 3 infants) than in the treatment success group (each 3% = 3 infants). To rule out a systematic effect due to the change in the use of the Griffiths to the Bayley scales during the study period, we analysed the data subsets of each test separately. In the Griffiths DQ 79 infants of the treatment success group (median 92 points; Q 1/3: 83/100) and 23 infants of the treatment failure group (median 92 points; Q 1/3: 89/96, p N 0.16) were tested at 22 months. In the Bayley test 32 infants of the treatment success group (median 90 points; Q 1/3: 75/99) and 13
89
Table 1 Socioeconomic characteristics of mothers, perinatal, PDA treatment and neonatal morbidity data summarized according to therapy success and failure. COX inhibitor therapy Success (128 infants) Socioeconomic data Age Level of education, n (%) No school certificate Nine school years Ten–twelve school years University qualification Unknown Single parent, n (%) Perinatal data Birthweight (g) GA (wks d) CRIB score Female, n (%) Completed, n (%) Prenatal steroids, n (%) Cesarean delivery, n (%) Cord blood pH RDS N 2°, n (%) Surfactant, n (%) Mech.ventilation, n (%) Ventilator days (ventilated infants only) FiO2 ≥ 0.30, n (%)
Neonatal morbidity NEC, n (%) ROP N 2°, n (%) Suppl. O2 wk. 36, n (%) IVH N 2°, n (%) PVL, n (%) Hospitalization days Survivors, n (%)
29 (24.5–33.5)
27 (23–32)
25 18 54 30 1 17
4 (7) 10 (19) 23 (43) 15 (28) 2 (4) 7 (13)
883 26.4 5 52 89 107 96 7.3 40 88 65 8
PDA treatment data First intervention (dol) Exposition to PDA (d) Reopening by infection, n (%)
Failure (54 infants)
(19) (14) (42) (23) (1) (13)
(770–1149) (25.5–28.1) (2–7) (41) (70) (84) (75) (7.25–7.34) (32) (69) (51) (4–18)
750 (665–881) 25.0 (24.3–26.5) 7 (4–10) 24 (44) 35 (65) 45 (83) 42 (78) 7.3 (7.24–7.35) 30 (58) 47 (87) 41 (76) 18 (12–26)
p
0.06 0.16
0.95
0.001 0.001 0.001 0.6 0.5 0.6 0.6 0.3 0.001 0.01 0.003 0.001
41 (32)
23 (43)
0.2
4 (3–6) 4 (3–5) 15 (12)
4 (2–7) 8 (6–11) 18 (33)
0.9 0.001 0.001
5 (10) 9 (17) 31 (57) 5 (10) 6 (11) 101 (82–131) 45 (83)
0.95 0.001 0.001 0.48 0.18 0.001 0.03
16 5 27 14 7 78 120
(13) (4) (21) (11) (6) (61–96) (94)
Median and quartiles are displayed, if not otherwise indicated. GA = gestational age, dol = day of life, RDS = respiratory distress syndrome, IVH = intraventricular hemorrhage, NEC = necrotizing enterocolitis, ROP = retinopathy of prematurity, PVL = cystic periventricular leukomalacia, and survivors = discharged home after primary hospital stay.
infants of the treatment failure group (median 80 points; Q 1/3: 49/87, p N 0.06) were tested at 24 months. No differences were found in hearing and walking ability or composite poor outcome at 24 months CA (Table 2).
Table 2 Follow-up data summarized according to therapy success and failure. Followed at two years
Hearing level N 35 dB, n (%) Hearing aid Blind Free walking at 2 years, n (%) CP Griffiths DQ 22 months b 88 or Bayley MDI 24 months b 70, n (%) Composite poor outcome
COX inhibitor therapy Success (109 infants)
Failure (32 infants)
p
11(10) 1 (1) 1 (1) 103 (95) 9 (5) 31 (29)
3 (9) 1 (1) 1 (1) 30 (94) 5 (16) 7 (22)
0.8 – – 0.70 0.24 0.50
37 (34)
8 (25)
0.40
Only data of infants with complete follow-up data are shown.
90
C. Rheinlaender et al. / Early Human Development 86 (2010) 87–92
3.3. Ibuprofen vs. indomethacin
Table 4 Follow-up data of both COX inhibitor therapy groups.
Comparison of infants who were treated with ibuprofen or indomethacin revealed no differences in their baseline clinical profile apart from a higher number of female individuals in the indomethacin group. No differences were found in perinatal and early and late morbidity data (Table 3). The number of infants with therapy related oliguria (indomethacin 3 and ibuprofen 1) and thrombocytopenia (indomethacin 16 and ibuprofen 6) was small in both groups. PDA closure rates and neurodevelopmental outcome at two years were alike in both groups (Tables 3 and 4). 4. Discussion
Table 3 Socioeconomic characteristics of mothers, perinatal, PDA treatment and neonatal morbidity data of both COX inhibitor therapy groups. COX inhibitor therapy Indomethacin (89 patients)
Perinatal data Birthweight (g) GA (wks d) CRIB score Female, n (%) Prenatal steroids, n (%) Completed, n (%) Cesarean delivery, n (%) Cord blood pH RDS N 2°, n (%) Surfactant, n (%) Mech.ventilation, n (%) Ventilator days (ventilated infants only) FiO2 ≥ 0.30, n (%)
Hearing level N 35 dB, n (%) Hearing aid Blind Free walking at 24 months, n (%) CP Griffiths DQ 22 months b 88 or Bayley MDI 24 months b 70, n (%) Composite poor outcome
COX inhibitor therapy Indomethacin (71 infants)
Ibuprofen (70 infants)
p
7 (10) 0 (0) 1 (1) 67 (94) 8 (11) 23 (32)
7 2 1 67 6 15
(10) (3) (1) (96) (9) (22)
0.81 – – 0.71 0.65 0.14
27 (38)
19 (28)
0.17
Only data of infants with complete follow-up data are shown.
The question of the impact of PDA ligation in preterm infants has been thoroughly discussed recently, as has the issue of the preferred drug. In an earlier study on the short-term complications of indomethacin treatment and PDA ligation, we proposed that surgical ligation should be reserved for infants not responding to pharmacological PDA closure [19]. With our current study, we aimed to add further information to this discussion by a detailed data analysis of our
Socioeconomic data Age Level of education, n (%) No school certificate Nine school years Ten–twelve school years University qualification Unknown Single parent, n (%)
Followed at two years
Ibuprofen (93 patients)
27 (24–33)
29 (24–33)
12 (14) 19 (21) 35 (39) 22 (25) 1 (1) 10 (11)
17 9 42 23 2 14
845 (730–1072) 26.0 (25.0–28.0) 6 (2–9) 44 (49) 70 (79) 62 (70) 68 (76) 7.3 (7.26–7.34) 36 (40) 67 (75) 54 (62) 15 (7–21)
850 26.2 5 32 82 62 70 7.3 34 68 52 15
p
0.42 0.23
(18) (10) (45) (25) (2) (15)
0.45
(717–1127) (25.0–27.6) (3–8) (34) (88) (67) (75) (7.23–7.34) (37) (73) (57) (4–11)
0.70 0.54 0.35 0.04 0.03 0.57 0.96 0.66 0.19 0.74 0.17 0.17
33 (37)
31 (33)
0.60
PDA treatment data First intervention (dol) Exposition to PDA (d) Reopening by infection, n (%) Treatment failure, n (%)
4 (3–7) 4 (4–6) 17 (19) 24 (27)
3 4 12 30
(2–6) (3–9) (13) (32)
0.65 0.47 0.24 0.44
Neonatal morbidity NEC, n (%) ROP N 2°, n (%) Suppl. O2 wk. 36, n (%) IVH N 2°, n (%) PVL, n (%) Hospitalization days Survivors, n (%)
10 (11) 9 (10) 31 (35) 10 (11) 7 (8) 89 (72–106) 80 (90)
11 5 27 9 6 76 85
(12) (5) (29) (10) (7) (60–96) (91)
0.20 0.21 0.39 0.47 0.71 0.07 0.91
Median and quartiles are displayed, if not otherwise indicated. GA = gestational age, dol = day of life, RDS = respiratory distress syndrome, IVH = intraventricular hemorrhage, NEC = necrotizing enterocolitis, ROP = retinopathy of prematurity, PVL = cystic periventricular leukomalacia, and survivors = discharged home after primary hospital stay.
hsPDA patient population. The flaws and disadvantages of the historical design are partly counterbalanced by strict therapy standards and examining staff that remained largely unchanged during the study period as well as a low rate of infants lost to followup. They are also mitigated by the fact that, with the exception of a change in preferred drug treatment, general management strategies for PDA did not undergo significant changes at our clinic in the specified time period, leading us to conclude that their effect on our data is inconsequential. The change from Griffiths DQ to Bayley scales for neurodevelopmental testing in accordance with international standards during the study period is one limitation of our study. Since the numerical scores of the two test sets are not equivalent, it is not possible to compare median test results directly. No comparative study of both tests has been done for preterm infants. It has, however, been demonstrated in other patient populations that both tests can be used interchangeably [20,21]. In the period where both test methods were used simultaneously, the outcome of 11 infants was equally categorized as either normal (7 infants) or poor (4 infants). The separate analysis of the median Griffiths and Bayley test results also revealed no statistical differences, though the significance of this is limited due to the small number of infants tested. The use of the same examiners over the whole study period is an advantage that allows us to consider our follow-up analysis reliable and comparable despite the use of two different tests. In our study, failed PDA closure following COX inhibitor therapy was associated with prematurity, neonatal morbidity and mortality, but not with poor neurodevelopmental outcome of the surviving infants at two years CA. Both the Schmidt and Clyman groups have published prospective studies on that issue [2,3]. The perinatal data of their and our study have consistently shown that the smallest, most premature infants are those that require surgical ligation after failed PDA closure by COX inhibitor therapy. In a retrospective study Noori's group has shown that the failure of ductal closure is associated with an increase in mortality in very preterm infants [22]. Failed PDA closure is associated with morbidity represented by prolonged hospitalization and higher mortality in our study population as well. Not surprisingly more infants in the therapy failure group were mechanically ventilated, given that respiratory setback (reintubation, FiO2 N 30%) was one criteria for the diagnosis of a hsPDA. Our echocardiographic data showed equal haemodynamic stress by PDA before intervention. PDA duration was longer in the therapy failure group supposedly because of the time needed to arrange for surgical ligation. The higher rate of reopening due to infection is probably caused by the immaturity and hence relative unresponsiveness of the PDA to COX inhibitors and its additional exposition to higher COX activity and prostaglandins. An efficacy of 60% closure rate after COX inhibitor therapy leaves room for improvement, as a higher success rate would probably lower ligation rate and hence could lead to better outcome results.
C. Rheinlaender et al. / Early Human Development 86 (2010) 87–92
Prolongation of COX inhibitor therapy as well as the escalation of doses has been discussed and investigated recently [23–25]. In our study a prolonged indomethacin protocol was used with 0.1 mg/kg up to six days but did not improve the rate of successful PDA closure in comparison to the ibuprofen group treated with the standard dosing of 10–5–5 mg/kg at 24 h intervals. Sperandio's group followed an interesting protocol of indomethacin dose titration up to 1 mg/kg in nonresponders after a conventional 3-dose course with 0.2 mg/kg indomethacin every 12 h until ductal closure was achieved in 98% of the infants without severe side effects [23]. Unfortunately, this retrospective analysis provided no information on the ROP rate. Clyman's group detected an association between the incidence of moderate and severe ROP and high indomethacin doses of 0.2 or 0.5 mg/kg for an extended 3-day course without a significant decrease in the rate of PDA in a multicenter randomised controlled trial, no recommendation was given on higher indomethacin doses for the prolonged treatment of PDA [24]. Overmeire's group investigated the association between areas under the curve for three conventional ibuprofen doses and successful PDA closure in VLBW infants and suggested a dosing scheme up to 18, 9, 9 mg/kg depending on postnatal age [25]. However, these theoretical findings still need to be tested prospectively. The evident differences of clinical data of the therapy failure compared to the success group lead to the expectation that neurodevelopmental outcome might differ as well, but surprisingly, we saw no statistical difference in neurodevelopmental outcome at 24 months CA. This result was not due to death of the most severely impaired infants during their second year of life. All together 5 infants died during the follow-up period, one in the therapy success group (persistent pulmonary hypertension) and 4 infants in the therapy failure group (1 diagnosis unknown, 2 infants with unknown syndrome, and 1 SIDS). The inclusion of death as a variable to composite poor outcome did not result in a significant difference. We can only presume that some infants caught up in their developmental delays. Furthermore, birth weight and CRIB score have a higher predictive power for neonatal morbidity and mortality than for neurodevelopmental outcome [26]. In the therapy failure group the follow-up recall rate is significantly lower than in the success group, 78% vs. 92% respectively. The fact that poor outcome is associated with parental incompliance for different reasons is nothing new — we should therefore not rule out the possibility that a higher follow-up rate would change outcome results, as the Clyman work has shown [3]. Additionally an extended follow-up assessment at 5 years corrected age will contribute to this question and might identify significant differences which are not apparent at two years. It seems almost impossible to separate the effect of surgical ligation from that of perinatal morbidity on the outcome of preterm infants. The group of Schmidt found that surgical PDA ligation in ELBW infants is associated with BPD, ROP and neurosensory impairment and suspected it of being a cause of a poor outcome at 18 months CA [2]. However, the meta-analysis of Fowlie and Davis did not show improved neurodevelopmental outcome despite a reduction in the rate of ligation in ELBW infants with prophylactic indomethacin therapy [1]. We also found a significantly higher rate of BPD and ROP in infants who underwent ligation after failed pharmacological PDA intervention. The Clyman group with a study population twice as large as ours demonstrated no effect of surgical ligation on neurodevelopmental impairment, but showed an independent effect on chronic lung disease [3]. We agree with the Brooks group that a randomised controlled trial of surgical PDA ligation after failed pharmacological treatment is required to clarify the role of this intervention on neurodevelopmental outcome in VLBW infants [27]. Until now no randomised controlled trial has provided data on neurodevelopmental follow-up at two years after therapeutic use for either COX inhibitor. In our study neurodevelopmental outcome did not differ after the use of either indomethacin or ibuprofen for
91
pharmacological closure of a hsPDA in VLBW infants. Several comparative studies account for equal efficacy of indomethacin and ibuprofen in closing a PDA, but the two COX inhibitors differ slightly in their side effects [4,28]. Indomethacin is known to cause higher decreases in renal and cerebral blood flow, and ibuprofen use has been associated with pulmonary hypertension, higher total bilirubin concentration and CLD [7,28,29]. At any rate, these different pharmacological profiles and the varied undesirable side effects associated with each drug, the major reason neonatologists switched to ibuprofen some years ago, did not seem to have relevant impact on neurodevelopmental outcome in our study population. Since we detected no significant differences in neurodevelopmental outcome at two years between the two COX inhibitor therapy groups, this retrospective data analysis does not indicate whether indomethacin or ibuprofen is preferable for the pharmacological therapy of a PDA. 5. Conclusion For the first time we were able to demonstrate that treatment for a hsPDA in preterm infants with either indomethacin or ibuprofen does not result in a difference in mortality or severely impaired neurodevelopmental outcome assessed at 2 years CA. Immaturity and mortality are associated with COX inhibitor treatment failure and hence ligation. Acknowledgements This study was supported by Orphan Europe and a Rahel-Hirschresearch grant (Dr. Koehne). We thank Boris Metze for the technical and statistical support. We thank Sarah Smithson-Compton for the suggestions regarding the English language. None of the authors has to declare a conflict of interest. References [1] Fowlie PW, Davis PG. Prophylactic intravenous indomethacin for preventing mortality and morbidity in preterm infants. Cochrane Database Syst Rev 2002;3: CD000174. [2] Kabra NS, Schmidt B, Roberts RS, Doyle LW, Papile L, Fanaroff A, et al. Neurosensory impairment after surgical closure of patent ductus arteriosus in extremely low birth weight infants: results from the Trial of Indomethacin Prophylaxis in Preterms. J Pediatr 2007;150:229–34. [3] Chorne N, Leonard C, Piecuch R, Clyman RI. Patent ductus arteriosus and its treatment as risk factors for neonatal and neurodevelopmental morbidity. Pediatrics 2007;119:1165–74. [4] Van Overmeire B, Chemtob S. The pharmacologic closure of the patent ductus arteriosus. Semin Fetal Neonatal Med 2005;10:177–84. [5] Couser RJ, Hoekstra RE, Ferrara TB, Wright GB, Cabalka AK, Connett JE. Neurodevelopmental follow-up at 36 months' corrected age of preterm infants treated with prophylactic indomethacin. Arch Pediatr Adolesc Med 2000;154(6): 598–602. [6] Aranda JV, Thomas R. Systematic review: intravenous Ibuprofen in preterm newborns. Semin Perinatol 2006;30:114–20. [7] Thomas RL, Parker GC, Van Overmeire B, Aranda JV. A meta-analysis of ibuprofen versus indomethacin for closure of patent ductus arteriosus. Eur J Pediatr 2005;164:135–40. [8] Safety Survey Report Named Patient Use of Ibuprofen Orphan Europe (PEDEA) Raport IBU/SURVEY/April 2003. [9] Lago P, Bettiol T, Salvadori S, Pitassi I, Vianello A, Chiandetti L, et al. Safety and efficacy of ibuprofen versus indomethacin in preterm infants treated for patent ductus arteriosus: a randomised controlled trial. Eur J Pediatr 2002;161:202–7. [10] Morgan AM, Aldag JC. Early identification of cerebral palsy using a profile of abnormal motor patterns. Pediatrics 1996;98(4 Pt 1):692–7. [11] Brandt I, Sticker E. Griffith Entwicklungsskalen (GES) zur Beurteilung der Entwicklung in den ersten beiden Lebensjahren. 2. erweiterte und überarbeitete Version. Göttingen: Beltz Test-GmbH; 2001. [12] Bayley N. Bayley scales of infant development (Bsid-II). 3rd ed. Sydney: Pearson Psychcorp; 1999. [13] Jobe AH, Bancalari E. Bronchopulmonary dysplasia. Am J Respir Crit Care Med 2001;163:1723–9. [14] Ehrenkranz RA, Walsh MC, Vohr BR, Jobe AH, Wright LL, Fanaroff AA, et al. Validation of the National Institutes of Health consensus definition of bronchopulmonary dysplasia. Pediatrics 2005;116:1353–60. [15] The International Committee for the Classification of the Late Stages of Retinopathy of Prematurity. An international classification of retinopathy of
92
[16]
[17]
[18]
[19]
[20] [21]
[22]
[23]
C. Rheinlaender et al. / Early Human Development 86 (2010) 87–92 prematurity. II. The classification of retinal detachment. Arch Ophthalmol 1987;105:906–12. Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr 1978;92:529–34. Resch B, Vollaard E, Maurer U, Haas J, Rosegger H, Müller W. Risk factors and determinants of neurodevelopmental outcome in cystic periventricular leucomalacia. Eur J Pediatr 2000;159:663–70. Surveillance of cerebral palsy in Europe: a collaboration of cerebral palsy surveys and registers. Surveillance of Cerebral Palsy in Europe (SCPE). Dev Med Child Neurol 2000;42(12):816–24. Koehne PS, Bein G, Alexi-Meskhishvili V, Wenig Y, Bührer C, Obladen M. Patent ductus arteriosus in very low birthweight infants: complications of pharmacological and surgical treatment. J Perinat Med 2001;29:327–34. Ramsay M, Fitzhardinge PM. A comparative study of two developmental scales: the Bayley and the Griffiths. Early Hum Dev 1977;1(2):151–7. Beail N. A comparative study of profoundly multiply handicapped children's scores on the Bayleys and Griffiths developmental scales. Child Care Health Dev 1985;11:31–6. Noori S, McCoy M, Friedlich P, Bright B, Gottipati V, Seri I, et al. Failure of ductus arteriosus closure is associated with increased mortality in preterm infants. Pediatrics 2009;123:e138–44. Sperandio M, Beedgen B, Feneberg R, Huppertz C, Brüssau J, Pöschl J, et al. Effectiveness and side effects of an escalating, stepwise approach to indomethacin
[24]
[25]
[26]
[27]
[28]
[29]
treatment for symptomatic patent ductus arteriosus in premature infants below 33 weeks of gestation. Pediatrics 2005;116:1361–6. Jegatheesan P, Ianus V, Buchh B, Yoon G, Chorne N, Ewig A, et al. Increased indomethacin dosing for persistent patent ductus arteriosus in preterm infants: a multicenter, randomized, controlled trial. J Pediatr 2008;153:183–9. Hirt D, Van Overmeire B, Treluyer JM, Langhendries JP, Marguglio A, Eisinger MJ, et al. An optimized ibuprofen dosing scheme for preterm neonates with patent ductus arteriosus, based on a population pharmacokinetic and pharmacodynamic study. Br J Clin Pharmacol 2008;65:629–36. Buhrer C, Grimmer I, Metze B, Obladen M. The CRIB (Clinical Risk Index for Babies) score and neurodevelopmental impairment at one year corrected age in very low birth weight infants. Intensive Care Med 2000;26(3):325–9. Brooks JM, Travadi JN, Patole SK, Doherty DA, Simmer K. Is surgical ligation of patent ductus arteriosus necessary? The Western Australian experience of conservative management. Arch Dis Child Fetal Neonatal Ed 2005;90:F235–9. Ohlsson A, Walia R, Shah S. Ibuprofen for the treatment of patent ductus arteriosus in preterm and/or low birth weight infants. Cochrane Database Syst Rev 2008;1: CD003481. Rheinlaender C, Helfenstein D, Walch E, Berns M, Obladen M, Koehne P. Total serum bilirubin levels during cyclooxygenase inhibitor treatment for patent ductus arteriosus in preterm infants. Acta Paediatr 2009;98:36–42.