Oa

Sb

Rabmi

Ccds

Lpostvapilr

orcis

t

ABCEELMP

Changes in Myocardial Function and Hemodynamics after Ligation of theDuctus Arteriosus in Preterm Infants

SHAHAB NOORI, MD, PHILIPPE FRIEDLICH, MD, MS EPI, ISTVAN SERI, MD, PHD, AND PIERRE WONG, MD

bjective To characterize the changes in systemic hemodynamics and systolic, diastolic, and global myocardial performancefter patent ductus arteriosus (PDA) ligation in very-low-birth-weight infants.

tudy design Echocardiograms were performed on 23 neonates (mean gestational age, 26.2 � 2.2 weeks) at 2.3 � 2.0 hoursefore PDA ligation (n � 23) and at 2.0 � 1.4 hours (n � 23) and 23.5 � 2.5 hours after (n � 11) PDA ligation.

esults Mean blood pressure, heart rate, load-independent contractility, shortening fraction, left ventricular (LV) afterload,nd diastolic function did not change. Preload (early and atrial mitral inflow velocities) decreased immediately after ligationut remained unchanged thereafter. LV output decreased and systemic vascular resistance increased after surgery. The LVyocardial performance index (MPI), a measure of global myocardial performance, deteriorated acutely after ligation but

mproved by 23.5 hours after surgery. Changes in LV MPI were most closely correlated with changes in LV output.

onclusions After PDA ligation, LV output and MPI decrease, due primarily to a decrease in LV preload, although LVontractility and diastolic function do not change. However, the changes in LV MPI after ligation also reflect an acuteeterioration followed by an improvement in global cardiac function, because LV loading conditions remained unchanged afterurgery and thus cannot explain the improvement in MPI by 24 hours after ligation. (J Pediatr 2007;150:597-602)

igation of the patent ductus arteriosus (PDA) has long been advocated in the premature neonate to abolish the potentiallydeleterious effects of the left-to-right ductal shunting on pulmonary function and outcome as well as on cerebral andgastrointestinal perfusion.1-4 However, little is known about the effects of PDA ligation on myocardial performance in

reterm neonates, and the available findings are contradictory.5,6 One study found a decrease in left ventricular (LV) cardiacutput and stroke volume after ligation without any associated changes in heart rate,5 whereas another study reported noignificant change in cardiac function but an increase in heart rate after PDA ligation.6 Ahird study described an increased requirement for vasopressor support in about 1/3 ofery-low-birth-weight (VLBW) infants undergoing PDA ligation.7 As for studies innimal models of PDA ligation, to date no study has evaluated postligation myocardialerformance after persistence of ductal patency for more than 1 week.8-10 Thus, becausen clinical practice PDA ligation is most often performed after the first week of postnatalife, animal studies have not addressed the question of how the neonatal myocardiumesponds to PDA ligation after a prolonged period of LV volume overload.

In older children, myocardial dysfunction has also been reported after transcathetercclusion or surgical closure of the PDA.11 However, there are major developmentallyegulated differences in myocardial structure and function between neonates and olderhildren or adults.12-14 For instance, the immature myocardium is more sensitive toncreases in afterload.15 This is an important point, because the sudden increase inystemic vascular resistance (SVR) after PDA ligation may acutely increase afterload.

The question of whether the immature myocardium of VLBW infants can adapt tohe sudden changes in systemic hemodynamics after PDA ligation has not been system-

Atrial mitral inflow DopplerP Blood pressureNICC Center for Newborn and Infant Critical Care

Early mitral inflow Dopplera Early mitral annulus tissue DopplerV Left ventricularPI Myocardial performance index

RV Right ventricularSF Shortening fractionSVR Systemic vascular resistanceVCFC Heart rate–corrected velocity of

circumferential fiber shorteningVLBW Very low birth weightVTI Velocity time integral

From the Divisions of Neonatal Medicine(S.N., P.F., I.S.) and Cardiology (P.W.), De-partment of Pediatrics, Childrens HospitalLos Angeles and Women’s and Children’sHospital, LAC � USC Medical Center,Keck School of Medicine, University ofSouthern California, Los Angeles, CA.

Supported in part by Emily and MichaelChasalow’s generous donation in supportof clinical research at the Center for New-born and Infant Critical Care, ChildrensHospital Los Angeles.Presented in part at the Pediatric AcademicSociety’s Annual Meeting, May 14-17, 2005,Washington, DC.

Submitted for publication Jul 21, 2006; lastrevision received Dec 29, 2006; acceptedJan 26 2007.

Reprint requests: Shahab Noori, MD, USCDivision of Neonatal Medicine, ChildrensHospital Los Angeles, 4650 Sunset Blvd, MS#31, Los Angeles, CA 90027. E-mail:snoori@chla.usc.edu.

0022-3476/$ - see front matter

Copyright © 2007 Mosby Inc. All rightsreserved.

DA Patent ductus arteriosus WS Wall stress

10.1016/j.jpeds.2007.01.035

597

atm

rt(tfsa

pqotcCsvac

dcfdt

vwppAstdmad

gtp22dt(

oeme

cSormcVztcpms

levrmpcuwsalmtt

uucrfwDBtrs(cpsmn

dv(cssa

5

tically studied to date. Consequently, we sought to charac-erize the changes in myocardial function and systemic he-odynamics in VLBW infants undergoing PDA ligation.

METHODSIn this observational study, data were collected both

etrospectively and prospectively. All VLBW infants admittedo the Center for Newborn and Infant Critical CareCNICC) at Childrens Hospital Los Angeles for PDA liga-ion between March 2004 and November 2005 were eligibleor the study. Between March 2004 and January 2005, theubjects were enrolled retrospectively; between February 2005nd November 2005, they were enrolled prospectively.

Neonates were enrolled if they were VLBW infants andresented with an echocardiographically confirmed PDA re-uiring surgical ligation according to the attending neonatol-gist and the consulting pediatric cardiologist. The decisiono perform ligation was based primarily on the presence ofardiovascular dysfunction and/or on the size of the PDA.ardiovascular dysfunction was defined by the presence of

ystemic hypotension not responding to low- to medium-doseasopressor/inotrope treatment and the presence of oliguriand/or metabolic acidosis. A PDA � 2 mm in size also wasonsidered hemodynamically significant.

All but 1 neonate was receiving mechanical ventilationuring the entire study period and had received at least 1ourse of indomethacin treatment. Patients were excludedrom this study if they had evidence of a hypertrophic orilated cardiomyopathy or had congenital heart defects otherhan a PDA or a patent foramen ovale.

The Childrens Hospital Los Angeles Institutional Re-iew Board approved the study. Informed consents wereaived by the Institutional Review Board for the retrospectivehase of the study and were obtained for all patients enrolledrospectively. In the CNICC at Childrens Hospital Losngeles, echocardiograms are performed routinely before and

hortly after PDA ligation. During the retrospective phase ofhe study, subjects were included only if a complete echocar-iographic assessment of cardiac function and systemic he-odynamics was available within 9 hours before and 6 hours

fter PDA ligation. A complete set of retrospective echocar-iographic data was available for 11 subjects.

During the prospective phase, complete echocardio-rams were performed in 12 patients within 6 hours preliga-ion and postligation and in 11 patients at 23.5 � 2.5 hoursostligation. Thus, this study presents hemodynamic data at.3 � 2.0 hours before and 2.0 � 1.4 hours postligation on all3 patients enrolled, and the findings of the third echocar-iogram performed at 23.5 � 2.5 hours postligation representhe findings of the prospectively enrolled preterm neonatesTable).

All echocardiograms were performed and analyzed by 1f the authors (S.N.), who is trained and certified in pediatricchocardiography, using a SONOS 5500 echocardiographyachine (Philips) equipped with 8- and 12-MHz transduc-

rs. On each echocardiogram, systolic, diastolic, and global c

98 Noori et al

ardiac function and hemodynamic features were evaluated.ystolic function was assessed by 2 load-dependent measuresf contractility—shortening fraction (SF) and heart rate-cor-ected velocity of circumferential fiber shortening (VCFC)—easured by M-mode15 and a load-independent measure of

ontractility known as stress-velocity index (the relation ofCFC to wall stress [WS]). The latter was calculated as ascore based on published normative data.16,17 The advan-

age of using the stress-velocity index is that it assesses myo-ardial contractility independent of influences from alteredreload and afterload. Therefore, while a load-dependenteasure of contractility assesses cardiac performance, the

tress-velocity index evaluates cardiac performance potential.LV and right ventricular (RV) outputs were also fol-

owed. The LV output was calculated using the aortic diam-ter and velocity time integral (VTI) measured at the aorticalve annulus from parasternal long-axis and apical views,espectively.18 The RV output was calculated using the pul-onary valve annulus diameter and VTI measured at the

ulmonary valve annulus from parasternal long-axis and sub-ostal views, respectively.18 Cardiac output was calculatedsing the following formula: (�D2/4) � VTI � heart rate,here D represents the diameter of the valve annulus. Dia-

tolic function was assessed by mitral valve inflow Dopplernd mitral annulus tissue Doppler performed at the septal andateral mitral annulus. The ratio of early (E) to atrial (A)

itral inflow Doppler (E/A) and early mitral inflow Dopplero early (Ea) mitral annulus tissue Doppler (E/Ea) were usedo evaluate diastolic function.19-21

Global myocardial function of the LV was assessedsing the myocardial performance index (MPI).22-24 A widelysed measure of global myocardial function,25-29 MPI is cal-ulated by dividing the sum of isovolumic contraction andelaxation times with the ejection time. Accordingly, theollowing formula was used to calculate MPI: (A � B)/B,here A is the time span between the end of one mitral flowoppler envelope to the beginning of the next envelope andis the ejection time.21 Because the MPI is inversely related

o myocardial function, an increase in MPI indicates a dete-ioration of global myocardial function.22-24 Preload was as-essed by measuring the LV internal diameter in diastoleLVIDD) and the mitral E and A velocities. SVR was cal-ulated using the following formula: SVR � (mean bloodressure – right atrial pressure)/LV output. Right atrial pres-ure was estimated as 4 mm Hg in all studies. PDA size waseasured as the diameter of the ductus arteriosus at its

arrowest site using color flow Doppler imaging.Blood pressure (BP) was measured directly by a trans-

ucer connected to an indwelling arterial catheter, or nonin-asively using the oscillometric method. Mean arterial BPrecorded at the time of the echocardiogram) was used toalculate WS. We used mean BP because it is more repre-entative of the entire cardiac cycle, and previous studies havehown an excellent correlation between mean arterial pressurend end-systolic pressure.17 Mean BP was also used to cal-

ulate SVR. As for systolic and diastolic BP, we present the

The Journal of Pediatrics • June 2007

aatp

S

s1c(sc

wMa22

(a

asp.rirvivTpPvpcB

T

SDMDMSVWSLRHLEAEEEEELLLS

EsPPP*

C

verage values over 6-hour blocks (the 6 hours before andfter ligation and the 6 hours from 18 to 24 hours postliga-ion). Because this was an observational study, all aspects ofatient care were directed by the attending neonatologist.

tatisticsAll data are given as mean � standard deviation unless

tated otherwise. Cardiovascular features were analyzed using-way analysis of variance for repeated-measures and pairwiseomparisons with adjustment for multiple comparisonsScheffe’s method). Univariate and multivariate linear regres-ion models were used as appropriate. A P value � .05 wasonsidered to indicate statistical significance.

RESULTSThe study population comprised 23 VLBW infants,

ith 11 enrolled retrospectively and 12 enrolled prospectively.ean birth weight was 845 � 280 g, and mean gestational

ge was 26.2 � 2.2 weeks. PDA ligation was done on day0 � 11 of postnatal life. Echocardiograms were performed

able. Changes in cardiac function and hemodynami

Preligation(�2.3 hours)

Postlig(2 ho

n n

ystolic BP (mm Hg) 23 51.9 � 9.4 23 49iastolic BP (mm Hg) 23 25 � 4.8 23 27.2ean BP (mm Hg)* 23 36 � 7 23 37opamine (�g/kg/min)* 23 8 � 8.2 23 7.5ean airway pressure (cm H2O)* 22 9.5 � 1.7 22 9.5

F (%) 23 38 � 7 23 33CFC (Circ/S) 23 1.37 � 0.25 23 1.40S (g/cm2) 23 18 � 8 23 18

tress-velocity index (z score) 23 �0.4 � 1 23 �0.6VO (mL/kg/min) 23 335 � 106 23 222VO (mL/kg/min) 22 372 � 108 22 343eart rate (bpm) 23 157 � 8 23 156VIDD (cm) 23 1.4 � 0.26 23 1.24(cm/s) 21 53 � 13 20 38(cm/s) 21 61 � 11 20 45

a, septal (cm/s) 18 4.2 � 1 18 3.3a, lateral (cm/s) 18 3.9 � 1.5 17 3.4/A 20 0.88 � 0.18 20 0.86/Ea, septal 17 12.8 � 3 16 12.4/Ea, lateral 17 15.8 � 7 15 13.4V MPI 23 0.26 � 0.10 23 0.53V IVCT� IVRT (ms) 23 44 � 13 23 78V ET (ms) 23 172 � 18 23 151VR (mm Hg/kg/min) 23 109 � 39 23 167

T, ejection time; IVCT, isovolumic contraction time; IVRT, isovolumic relaxation time;ignificant; RVO, right ventricular output.

value1 � Preligation (�2.3 hours) versus postligation (2 hours).value2 � Preligation (�2.3 hours) versus postligation (23.5 hours).value3 � Postligation (2 hours) versus postligation (23.5 hours).

Values at the time of echocardiography.

.3 � 2.0 hours before ligation (n � 23) and 2.0 � 1.4 hours d

hanges in Myocardial Function and Hemodynamics after Ligation of the

n � 23) and 23.5 � 2.5 hours (n � 11) after ligation. Theverage size of the PDA was 2.5 � 0.9 mm.

Changes in cardiovascular measures and hemodynamicsre given in the Table. Compared with preligation status, aignificant reduction in LV output occurred both immediatelyostligation (P � .0001) and at 23.5 hours postligation (P �

02). Because there was no significant change in heart rate, theeduction in LV output resulted principally from a reductionn LV preload. Although the decrease in LVIDD did noteach statistical significance, the drop in mitral E and Aelocities (both P � .0001) indicates a reduction in preloadmmediately after ligation. LVIDD and mitral E and Aelocities remained unchanged at 23.5 hours postligation.here were no significant changes in RV output, mean airwayressure, and dopamine dose before and immediately afterDA ligation; however, in 35% of the patients (n � 8), theasopressor dose was increased by at least 25% by 23.5 hoursostligation. Systolic and mean BP did not significantlyhange after PDA ligation, but a modest increase in diastolicP occurred by 23.5 hours postligation (P � .036).

As for the measures of systolic function, SF and VCFC

Postligation(23.5 hours) P

value1P

value2P

value3 ANOVAn

.8 23 51.1 � 9.8 NS NS NS NS

.1 23 28.7 � 5.5 NS 0.036 NS 0.0411 40 � 10 NS NS NS NS

.5 11 7.8 � 11.9 NS NS NS NS

.8 10 9.7 � 2.3 NS NS NS NS11 36 � 8 NS NS NS NS

.41 11 1.39 � 0.4 NS NS NS NS11 18 � 8 NS NS NS NS

.9 11 �1.1 � 1.8 NS NS NS NS6 11 246 � 77 .0001 .02 NS 0.000113 11 345 � 109 NS NS NS NS1 11 158 � 11 NS NS NS NS.25 11 1.25 � 0.23 NS NS NS NS

11 44 � 12 .0001 NS NS 0.00052 11 52 � 12 .0001 NS NS 0.0003.1 11 3.7 � 1.1 NS NS NS NS.2 10 4.1 � 2 NS NS NS NS.16 11 0.83 � 0.1 NS NS NS NS.8 11 12.2 � 2.8 NS NS NS NS.4 10 12.6 � 4.4 NS NS NS NS.19 11 0.39 � 0.09 .0001 .05 .03 0.00010 11 63 � 14 .0001 .007 .05 0.00018 11 162 � 15 .0001 NS NS 0.00059 11 148 � 36 .0001 NS NS 0.0005

, left ventricular internal diameter at end-diastole; LVO, left ventricular output; NS, not

cs

ationurs)

� 9� 4� 8� 7� 1� 8� 0� 9� 1� 6� 1� 1� 0� 9� 1� 1� 1� 0� 3� 4� 0� 2� 1� 5

LVIDD

id not change. The load-independent measure of myocardial

Ductus Arteriosus in Preterm Infants 599

crm

arotr2dr.sl

[asbeo

datownMaorapLTrMtwcmwss

Pmritag

ciMaatwv

vnMaptflsafiTtulfiaufbst

Lwttgc

eacpeoitspamfcp

6

ontractility—the stress-velocity index—was unchanged andemained within 2 standard deviations below the normativeean (z score).

Although E and A significantly decreased immediatelyfter PDA ligation, both the mean E/A and E/Ea ratiosemained unchanged throughout the study. LV MPI deteri-rated immediately after ligation (P � .0001); however, al-hough LV MPI remained elevated in the postoperative pe-iod, an improvement was seen by 23.5 hours relative to the-hour postligation value (P � .03). The LV ejection timeecreased, and the sum of the isovolumic contraction andelaxation times increased immediately after ligation (P �0001). Although WS remained unchanged throughout thetudy period, SVR increased significantly immediately afterigation (P � .0003).

Among the determinants of LV output (ie, LVIDDpreload], stress-velocity index [contractility], WS [afterload],nd heart rate) and MPI, changes in LVIDD and MPI wereignificantly correlated with LV output 2 hours after ligationy univariate analysis. When adjusted for all variables, how-ver, only MPI remained significantly correlated with LVutput (R2 � .52; P � .02).

Because we observed a decrease in LV output andeterioration in MPI immediately after ligation, we furthernalyzed the data to investigate correlations between any ofhe preligation factors and the postligation changes in LVutput and MPI. The factors examined included infant’s birtheight and gestational age, mother’s postmenstrual age, post-atal age at ligation, PDA size and preligation LV output,PI, stress velocity index, E/A ratio, heart rate, SF, LVIDD,

nd pressor/inotrope level. Only preligation PDA size, LVutput, and stress velocity index exhibited a significant cor-elation with LV output at 2 hours postligation by univariatenalysis. When adjusted for all of these 3 variables, only thereligation PDA size correlated significantly with changes inV output at 2 hours postligation (R2 � .58; P � .0008).his finding indicates that the larger the PDA, the greater the

eduction in LV output after ligation. As for the postligationPI, univariate analysis revealed a significant direct correla-

ion with preligation PDA size and an inverse correlationith preligation MPI. These correlations remained statisti-

ally significant even after adjustment for the other variable byultivariate analysis (R2 � .49; P � .001). No differencesere seen in the patterns of changes in cardiac function and

ystemic hemodynamics between the prospective and retro-pective data sets (data not shown).

DISCUSSIONIn our study, LV MPI increased in all but 1 patient after

DA ligation. Because MPI is inversely related to globalyocardial function, an increase in MPI indicates a deterio-

ation of cardiac function. However, because contractility andndices of diastolic function remained unchanged after liga-ion, the deterioration of LV MPI suggests that MPI may beffected by loading conditions in preterm neonates or a

reater sensitivity of this index to changes in cardiac function t

00 Noori et al

ompared with more conventional measures. Indeed, studiesn animals have suggested at least some load dependence of

PI, because an acute reduction in preload is associated withn increase MPI.30 However, the relationship between MPInd preload also appears to be affected by cardiac func-ion.30,31 For example, volume loading does not alter the MPIith normal LV function, even though MPI decreases witholume loading if LV dysfunction exists.31

It is important to keep in mind that a PDA causes LVolume overload, with the load decreasing (and presumablyormalizing) after PDA ligation. In our patients, baseline LVPI was in the normal range (0.33 � 0.08)21 but increased

fter ligation to an abnormally high level. By 23.5 hoursostligation, however, LV MPI began to normalize evenhough the loading conditions remained unchanged. There-ore, although a sudden decrease in preload immediately afterigation likely contributed to the increase in LV MPI, the datauggest that transient deterioration in global cardiac functionlso occurred, possibly due to the acute decrease in myocyteber length after the sudden decrease in LV loading volume.he finding that PDA size correlates with both the postliga-

ion decrease in LV output and the deterioration in LV MPInderscore the significance of the severity of volume over-oading on postligation myocardial performance. Thus, thesendings indicate that the decrease in stroke volume immedi-tely after ligation was caused by both the acute LV volumenloading and cardiac function deterioration. Global cardiacunction then improved by 23.5 hours postligation, perhapsecause the normalized and unchanged volume loading afterurgery promoted a “resetting” of myocyte fiber length overime.

Furthermore, multivariate analysis revealed that changes inV output after ligation correlated with changes in MPI but notith any of the determinants of LV output (preload, contrac-

ility, heart rate, and afterload). This finding further supportshe notion that MPI might be a more sensitive measure oflobal myocardial function in preterm neonates than theonventional methods.

There is little information in the literature regarding theffect of PDA ligation on myocardial performance. Mostnimal studies have been limited to describing changes inardiac function after closure of the PDA within the first 24ostnatal hours10,11 and thus do not represent the frequentlyncountered clinical situation in which the PDA remainspen for many days or weeks before ligation. Therefore, thempact of prolonged exposure of the immature myocardiumo a high preload cannot be extrapolated from these animaltudies. One recent study reported the effect of PDA ligationerformed on postnatal day 6 on cardiopulmonary function inpreterm baboon model.9 Although the findings suggest aedium-term beneficial effect of PDA ligation on myocardial

unction, the study did not systematically examine thehanges in cardiac function in the immediate postoperativeeriod.

As for human data on cardiac function after ligation,

he information is equally scarce. To the best of our knowl-

The Journal of Pediatrics • June 2007

epmatsafrmissdp

bPcmhtpt

pdibufMvsritiasu

figboavomfwd

P

pmnmipf

1rb2Op3I14m5a16li7p8eR9Lo1Lb1uA11ou1p1H1p1dC1E1fd2dbD2Ic2om1

C

dge, only 2 studies have specifically addressed this issue inreterm infants. Lindner et al5 assessed cardiac function byeasuring LV output, stroke volume, and heart rate 2 days

fter PDA ligation and compared the findings with preliga-ion measurements. Similar to our observations, they found aignificant decrease in LV output and stroke volume withoutsignificant change in heart rate. In contrast, Kimball et al6

ound an increase in systemic vascular resistance and heartate and no changes in the load-independent measures ofyocardial contractility, LV output, or afterload. Interest-

ngly, although the decrease in LV output did not reachtatistical significance,6 the magnitude of the decrease wasimilar to what we found. The lack of a statistically significantecrease in LV output may be related to the small number ofatients enrolled in the study.6

Changes in RV output after PDA ligation have noteen reported. We found no difference in RV output afterDA ligation. In addition, we found a higher RV outputompared with LV output throughout our study. This findingay be explained by the observation that all of our patients

ad a left-to-right or a predominantly left-to-right shunt athe level of the foramen ovale. Indeed, in the presence of aatent foramen ovale, others have also found a greater RVhan LV output even with concomitant ductal shunting.32

During the first 24 hours after ligation, 35% of ouratients required at least a 25% increase in the dose ofopamine to maintain BP in the normal range. Indeed, an

ncreased vasopressor requirements after PDA ligation haseen reported in approximately 1/3 of patients who havendergone surgical PDA ligation.7 Among the various cardiacunction indices evaluated in our patients (including the

PI), none was found to predict an increase in the need forasopressor support. Although this could be due to the smallample size, it is also conceivable that the increased dopamineequirement might have resulted from postoperative changesn vascular tone rather than cardiac function. Down-regula-ion of cardiovascular adrenergic receptors, relative adrenalnsufficiency, and anesthesia could be associated with alter-tions in vasomotor tone, necessitating escalation of vasopres-or support to maintain BP in critically ill preterm neonatesndergoing surgery.33,34

In addition to the limitations inherent to the use ofunctional echocardiography in the preterm neonate, anothermportant limitation of our study is the lack of a controlroup. Therefore, although echocardiograms were performedefore and after PDA ligation and each patient served as hisr her own control, we could not control for the effects ofnesthesia or the stress of surgery. There was some individualariation in the administered dopamine dose for each patientver the course of the first postoperative day. This variationight have contributed to some of the changes in the cardiac

unction measurements. However, for the study group as ahole, there was no significant difference in mean dopamineose at the time of each echocardiogram.

The acute decrease in LV output documented after

DA ligation might result from both a reduction in LV

2a

hanges in Myocardial Function and Hemodynamics after Ligation of the

reload and the deterioration of global cardiac function im-ediately after surgery. As the myocardium then resets to the

ew loading conditions, its performance improves, as docu-ented in our study. It is unclear whether the postligation

ncrease in vasopressor support is required due to abnormaleripheral vasoregulation, a subtle decrease in global cardiacunction, or a combination of these 2 factors.

REFERENCES. Cassady G, Crouse DT, Kirklin JW, Strange MJ, Joiner CH, Godoy G, et al. Aandomized, controlled trial of very early prophylactic ligation of the ductus arteriosus inabies who weighed 1000 g or less at birth. N Engl J Med 1989;320:1511-6.. Thibeault DW, Emmanouilides GC, Nelson RJ, Lachman RS, Rosengart RM,h W. Patent ductus arteriosus complicating the respiratory distress syndrome in

reterm infants. J Pediatr 1975;86:120-6.. Dykes FD, Lazzara A, Ahmann P, Blumenstein B, Schwartz J, Brann AW.ntraventricular hemorrhage: a prospective evaluation of etiopathogenesis. Pediatrics980;66:42-9.. Ryder RW, Shelton JD, Guinan ME. Necrotizing enterocolitis: a prospectiveulticenter investigation. Am J Epidemiol 1980;112:113-23.

. Lindner W, Seidel M, Versmold HT, Dohlemann C, Riegel KP. Stroke volumend left ventricular output in preterm infants with patent ductus arteriosus. Pediatr Res990;27:278-81.. Kimball TR, Ralston MA, Khoury P, Crump RG, Cho FS, Reuter JH. Effect ofigation of patent ductus arteriosus on left ventricular performance and its determinantsn premature neonates. J Am Coll Cardiol 1996;27:193-7.. Moin F, Kennedy KA, Moya FR. Risk factors predicting vasopressor use afteratent ductus arteriosus ligation. Am J Perinatol 2003;20:313-20.. McCurnin DC, Yoder BA, Coalson J, Grubb P, Kerecman J, Kupferschmid J,t al. Effect of ductus ligation on cardiopulmonary function in premature baboons. Am Jespir Crit Care Med 2005;172:1569-74.. Baylen BG, Ogata H, Ikegami M, Jacobs HC, Jobe AH, Emmanouilides GC.eft ventricular performance and regional blood flows before and after ductus arteriosuscclusion in premature lambs treated with surfactant. Circulation 1983;67:837-43.0. Taylor AF, Morrow WR, Lally KP, Kinsella JP, Gerstmann DR, deLemos RA.eft ventricular dysfunction following ligation of the ductus arteriosus in the pretermaboon. J Surg Res 1990;48:590-6.1. Galal MO, Amin M, Hussein A, Kouatli A, Al-Ata J, Jamjoom A. Left ventric-lar dysfunction after closure of large patent ductus arteriosus. Asian Cardiovasc Thoracnn 2005;13:24-9.

2. Anderson PA. The heart and development. Semin Perinatol 1996;20:482-509.3. Balaguru D, Haddock PS, Puglisi JL, Bers DM, Coetzee WA, Artman M. Rolef the sarcoplasmic reticulum in contraction and relaxation of immature rabbit ventric-lar myocytes. J Mol Cell Cardiol 1997;29:2747-57.4. Noori S, Seri I. Pathophysiology of newborn hypotension outside the transitionaleriod. Early Hum Dev 2005;81:399-404.5. Snider AR, Serwer GA, Ritter SB, Gersony RA. Echocardiography in Pediatriceart Disease. 2nd ed. St Louis: Mosby; 1997.

6. Rowland DG, Gutgesell HP. Noninvasive assessment of myocardial contractility,reload, and afterload in healthy newborn infants. Am J Cardiol 1995;75:818-21.7. Rowland DG, Gutgesell HP. Use of mean arterial pressure for noninvasiveetermination of left ventricular end-systolic wall stress in infants and children. Am Jardiol 1994;74:98-9.

8. Skinner J, Alverson D, Hunter S. Echocardiography for the Neonatologist.dinburgh: Churchill Livingstone; 2000.9. Schmitz L, Stiller B, Koch H, Koehne P, Lange P. Diastolic left ventricularunction in preterm infants with a patent ductus arteriosus: a serial Doppler echocar-iography study. Early Hum Dev 2004;76:91-100.0. Schmitz L, Stiller B, Pees C, Koch H, Xanthopoulos A, Lange P. Doppler-erived parameters of diastolic left ventricular function in preterm infants with airth weight �1500 g: reference values and differences to term infants. Early Humev 2004;76:101-14.

1. Eidem BW, McMahon CJ, Cohen RR, Wu J, Finkelshteyn I, Kovalchin JP, et al.mpact of cardiac growth on Doppler tissue imaging velocities: a study in healthyhildren. J Am Soc Echocardiogr 2004;17:212-21.2. Tei C, Ling LH, Hodge DO, Bailey KR, Oh JK, Rodeheffer RJ, et al. New indexf combined systolic and diastolic myocardial performance: a simple and reproducibleeasure of cardiac function. A study in normals and dilated cardiomyopathy. J Cardiol

995;26:357-66.

3. Eidem BW, Tei C, O’Leary PW, Cetta F, Seward JB. Nongeometric quantitativessessment of right and left ventricular function: myocardial performance index in

Ductus Arteriosus in Preterm Infants 601

n12ce2Pv2p2Ip2nc

2dh3Vi3t3s13t3

6

ormal children and patients with Ebstein anomaly. J Am Soc Echocardiogr 1998;1;849-56.4. Tei C, Nishimura RA, Seward JB, Tajiik J. Noninvasive Doppler-derived myo-ardial performance index: correlation with simultaneous measurement of cardiac cath-terization. J Am Soc Echocardiogr 1997;10;1969-78.5. Peltier M, Slama M, Garbi S, Enriquez-Sarano ML, Goissen T, Tribouilloy CM.rognostic value of Doppler-derived myocardial performance index in patients with leftentricular systolic dysfunction. Am J Cardiol 2002;90:1261-3.6. Bruch C, Schmermund A, Marin D, Katz M, Bartel T, Schaar J, et al. Tei index inatients with mild-to-moderate congestive heart failure. Eur Heart J 2000;21:1888-95.7. Leonard GT Jr, Fricker FJ, Pruett D, Harker K, Williams B, Schowengerdt KO Jr.ncreased myocardial performance index correlates with biopsy-proven rejection inediatric heart transplant recipients. J Heart Lung Transplant 2006;25:61-6.8. Mooradian SJ, Goldberg CS, Crowley DC, Ludomirsky A. Evaluation of a

oninvasive index of global ventricular function to predict rejection after pediatricardiac transplantation. Am J Cardiol 2000;86:358-60.

aw

02 Noori et al

9. Ichihashi K, Yada Y, Takahashi N, Honma Y, Momoi M. Utility of a Doppler-erived index combining systolic and diastolic performance (Tei index) for detectingypoxic cardiac damage in newborns. J Perinat Med 2005;33:549-52.0. Broberg CS, Pantely GA, Barber BJ, Mack GK, Lee K, Thigpen T, et al.alidation of the myocardial performance index by echocardiography in mice: a non-

nvasive measure of left ventricular function. J Am Soc Echocardiogr 2003;16:814-23.1. Lavine SJ. Effect of heart rate and preload on index of myocardial performance inhe normal and abnormal left ventricle. J Am Soc Echocardiogr 2005;18:133-41.2. Evans N, Iyer P. Assessment of ductus arteriosus shunt in preterm infantsupported by mechanical ventilation: effect of interatrial shunting. J Pediatr 1994;25:778-85.3. Seri I, Evans J. Controversies in the diagnosis and management of hypotension inhe newborn infant. Curr Opin Pediatr 2001;13:116-23.4. Ng PC, Lee CH, Lam CW, Ma KC, Fok TF, Chan IH, et al. Transient

drenocortical insufficiency of prematurity and systemic hypotension in very low birth-eight infants. Arch Dis Child Fetal Neonatal Ed 2004;89:F119-26.

50 Years Ago in The Journal of PediatricsSUICIDE IN CHILDREN AND ADOLESCENTS

Bakwin H. J Pediatr 1957;50:749-69

Bakwin sought to raise awareness about adolescent suicide, the relative importance of which “has increased as othercauses of death, notably the infections, have diminished.” Suicide was the fifth most frequent cause of death in youngadults in the United States (US) in 1957; it is now the third most common cause of death in youth ages 10 to 24 years.50 years ago suicide accounted for 2.5% of all deaths in youth 15 to 19 years old; in 2004, it accounted for 11.7% of alldeaths in those 10 to 24 years old. Then, as now, the most common method of suicide in adolescents was throughfirearms, although in the US during the past decade, suffocation has become the most common method in adolescents10 to 14 years old.

During the past few decades, the suicide rate has generally increased in adolescents. A half-century ago, suicide ratesvaried considerably across nations, ranging from 0.6 of every 100,000 male adolescents in Ireland to 26.1 of every 100,000male adolescents in Japan. The wide range (and male preponderance) in rates of countries still exists, but the range ishigher (from 2-44 per 100,000); the greatest rates are now found in Europe. However, in the US, there has been a declinein suicide rates, falling from 6.2 to 4.6 per 100,000 population in youth ages 10 to 19 years between 1991 and 2002.

The most notable difference in suicide today compared with that 50 years ago is the enormous attention focused onthe problem and the availability of preventive measures. Only 2 small paragraphs were needed to describe preventionoptions available at that time; hospitalization after a suicide attempt appeared to be helpful, and there was considerationof creating a society similar to Alcoholics Anonymous. Since then, hundreds of prevention programs have been developedand evaluated, with some evidence that school-based suicide prevention programs based on behavioral change andcoping, those based on skill training and social support, or both can be effective. A wide range of medical andnon-medical therapies have been developed, some of which have been demonstrated to be effective in specific conditions.The observational relationship between depression and subsequent suicide attempts (in 1957 and now) led to thewidespread interest in identifying and treating youth who are depressed. However, the possibility of an adverse effect ofsome anti-depressants illustrates the complexity of the disorders that result in the final common pathway of suicide.

Bonita Stanton, MDProfessor and Schotanus Family Endowed Chair of Pediatrics

Pediatrician-in-ChiefCarman and Ann Adams Department of Pediatrics

Children’s Hospital of MichiganWayne State University School of Medicine

Detroit, Michigan10.1016/j.jpeds.2007.02.003

The Journal of Pediatrics • June 2007