Journal of Pediatric Surgery (2005) 40, 1748–1752

www.elsevier.com/locate/jpedsurg

Effects of surgical repair of congenital diaphragmatic

hernia on cerebral hemodynamics evaluated by

near-infrared spectroscopy

Andrea Dotta*, Jole Rechichi, Francesca Campi, Annabella Braguglia,Sabrina Palamides, Irma Capolupo, Simona Lozzi, Alessandro Trucchi,Carlo Corchia, Pietro Bagolan, Marcello Orzalesi

Department of Medical and Surgical Neonatology–Bambino Gesu Children’s Hospital, 00165 Rome, Italy

0022-3468/$ – see front matter D 2005

doi:10.1016/j.jpedsurg.2005.07.001

T Corresponding author. Neonatal In

00165 Rome, Italy. Tel.: +39066859242

E-mail addresses: andreadotta@iol.i

Index words:Congenital diaphragmatic

hernia;

Near-infrared

spectroscopy;

Newborn

Abstract

Background: Cardiorespiratory stabilization is recommended before surgical repair of congenital

diaphragmatic hernia (CDH) because surgery may induce a transitory deterioration of chest compliance

and gas exchange. It is not known if surgical intervention can affect cerebral circulation and oxygenation.

Aim: The aim of the study was to assess noninvasively, by near-infrared spectroscopy, the possible

changes in cerebral hemodynamics and oxygenation associated with surgical repair of CDH.

Subjects: Twenty-five newborns with severe CDH (birth weight, 3057F 354 g; gestational age, 37.8F

1.8 weeks; male/female newborns, 15/10; left/right CDH, 19/6) were sedated, paralyzed, and

mechanically ventilated by conventional gentle ventilation and surgically corrected at a median age

of 2.7 days (min-max, 2-14 days) after cardiorespiratory stabilization.

Methods: Heart rate (HR [beats per minute]), preductal transcutaneous oxygen saturation (tcSao2 [%]),

carbon dioxide tension (tcPco2 [Torr]), and mean arterial blood pressure (mm Hg) were continuously

monitored. Inspired fractional oxygen concentration (Fio2) was adjusted to maintain a preductal tcSao2

of greater than 80%, whereas the ventilator’s settings were kept unchanged throughout the surgical

procedure. Cerebral hemodynamics was assessed by near-infrared spectroscopy (NIRO 300,

Hamamatsu Photonics, Japan), recording continuously and noninvasively the relative changes in

concentration of oxygenated (DO2Hb [lmol/L]), deoxygenated (DHHb [lmol/L]), and total (DtHb

[lmol/L]) hemoglobin; the tissue oxygenation index (TOI [%]) was also calculated (TOI = O2Hb/O2Hb

+ HHb). Total hemoglobin concentration is considered to be representative of cerebral blood volume.

Arterial blood gases were also measured at the beginning (T1) and at the end of surgery (T2). For all

measurements, results at T1 and at T2, as well as the differences between T1 and T2, have been

expressed as means or medians and SDs or 95% confidence intervals or ranges. The differences

between T1 and T2 were considered statistically significant for a P value of less than .05 by the Student

t test for paired values.

Results: At T1, mean tcSao2%was 94.1 % (SD, 4.6) with a Fio2 of 0.25 (SD, 0.1); at T2, to obtain similar

values of tcSao2 (93.4%; SD, 4.4), it was necessary to increase the Fio2 to 0.37 (SD, 0.14; P b .001).

Elsevier Inc. All rights reserved.

tensive Care Unit, Department of Medical and Surgical Neonatology, bBambino GesuQ Children’s Hospital–IRCCS

7; fax: +39066875540.

t, dotta@opbg.net (A. Dotta).

,

Cerebral hemodynamics in congenital diaphragmatic hernia 1749

Mean HR at T1 was 149.5 beats per minute (SD, 9.1) and increased significantly (P b .05) at T2 (165.2

beats per minute; SD, 14.2). Mean arterial blood pressure was 54.7 mm Hg (SD, 7.7) at T1 and did not

change appreciably at T2 (55.6 mm Hg; SD, 8.1). Moreover, tcPco2 did not change significantly

during the procedure (mean tcPco2 = 49.9 Torr [SD, 12.8] at T1 and 57.3 mm Hg [SD, 17.9] at T2).

O2Hb and tHb decreased (P b .001 and b.005) and HHb increased (P b .05) significantly during the

surgical procedure (mean D [SD]: DO2Hb= ÿ10.9 lmol/L [9.7], DtHb = ÿ7.5 lmol/L [11.7], and

DHHb = ÿ3.5 lmol/L [6.8]). Mean TOI was 70% at T1 (normal values N60%) and decreased

significantly at T2 (mean DTOI = ÿ6.1% [SD, 10.6]). In all infants, the greatest changes occurred when

the viscera were positioned into the abdomen.

Conclusions: Notwithstanding the initial cardiorespiratory stabilization, surgical repair of CDH was

associated with a rise in HR and oxygen requirement and a drop in cerebral tHb and O2Hb, suggesting a

reduction in cerebral blood volume and oxygenation. These events were probably due to the combined

effects of an increase in right to left shunting (as indicated by the increased oxygen requirement) and a

decrease in venous return (possibly due to compression of the inferior vena cava by the viscera positioned

into the abdomen). These preliminary results reinforce the importance of achieving a good

cardiorespiratory stability before undertaking surgical correction of CDH to minimize the possible

interference of the procedure with cerebral circulation and oxygenation.

D 2005 Elsevier Inc. All rights reserved.

Congenital diaphragmatic hernia (CDH) is still associ-

ated with a high mortality rate, ranging widely from 25% to

as much as 74% in different centers [1,2]. Among the

various strategies explored in the treatment of CDH, a

delayed approach to surgical repair, after a period of

cardiorespiratory stabilization and gentle ventilation, with

permissive hypercapnia, has been associated with lower

mortality rates [1,3]. Until 20 years ago, CDH was

considered a surgical emergency. In 1986, Cartlidge et al

[4] reported a marked improvement in survival in CDH

newborns who had been stabilized for 4 to 16 hours before

surgery when compared with a historical control group

(from 13% to 54%). Two years later, Langer et al [5]

showed similar mortality rates in CDH patients operated at

a mean age of 4 hours (42%) or at 24 hours (50%). In the

same years, Sakai et al [6] described a significant

deterioration in pulmonary compliance in the immediate

postoperative period in infants with CDH. Following these

studies, the need for immediate surgical intervention in

infants with CDH was reconsidered. In 1998, a report by

the Congenital Diaphragmatic Hernia Study Group, de-

scribing the surgical management of CDH in various

institutions in North America, Europe, and Australia,

highlighted that only 38% of surgical repairs were

performed within 24 hours, whereas in the majority of

cases surgery, surgical repairs was delayed [7]. More

recently, Boloker et al [8] treated 120 infants with high-

risk CDH using permissive hypercapnia/spontaneous respi-

ration/elective repair, with an excellent survival rate and

minimal pulmonary morbidity at discharge. In 2004,

Rozmiarek et al [9] concluded that the most important

negative factors influencing survival in CDH are associated

cardiac defects, renal failure, and initial blood gases and not

the timing of surgery.

Ventilatory strategies used to guarantee adequate gas

exchange in infants with CDH have changed in the last

decades as the understanding of neonatal pathophysiology

improved, showing that a significant cause of mortality

was iatrogenic injury of the hypoplastic lung. In 1981,

Drummond et al [10] showed that alkalosis and hypocapnia

could reverse ductal shunting; as a result, hyperventilation

became a popular ventilatory strategy for CDH in the

following years. More recently, based on the work of Wung

et al [11] and Langham et al [12], it has been recommended

to avoid high peak inspiratory pressures (PIPs) and lung

overdistension. However, the optimal mode of ventilation

for patients with CDH is still unclear. At present, a

permissive hypercapnia strategy is used in most of centers;

moreover, strategies designed to limit lung overdistension

and excessive inspiratory pressures (gentle ventilation) are

considered to be the most adequate modes of ventilation in

CDH [2].

The search for the best ventilatory and surgical strategies

for the management of patients with CDH has focused

mainly on the cardiorespiratory system, and little is known

concerning the effects of surgical repair on cerebral

hemodynamics. Surgery could affect cerebral circulation

in 2 ways: through changes in arterial carbon dioxide

tension (Paco2), which is considered a potent regulator of

the cerebral circulation [13], and through modifications in

arterial blood pressure and systemic venous return, second-

ary to compression of the inferior vena cava by the herniated

viscera repositioned into the abdomen.

Among the different techniques available for studying

cerebral hemodynamics and oxygenation, near-infrared

spectroscopy (NIRS), first introduced by Jobsis [14] in

1977, has gained rapid clinical acceptance because it is

noninvasive, well tolerated by the patient, and relatively

easy to apply at the bedside [15,16]. NIRS takes advantage

of the fact that near-infrared light, generated by a laser

beam, can penetrate up to 6 to 8 cm through the skull into

the neonatal brain. The light beam spreads diffusely into the

Table 1 Major characteristics of the 25 CDH infants studied

Gender (male/female) 15/10

Gestational age, wk (mean F SD) 37.8 F 1.8

Birth weight, g (mean F SD) 3057 F 354

Prenatal diagnosis (n) 21

Side of hernia (left/right) 19/6

Age at surgery, d (mean [median] F SD) 3.5 [2.7] F 2.5

A. Dotta et al.1750

brain; it is partially absorbed by various chromophores

(including oxygenated [O2Hb] and deoxygenated (HHb)

hemoglobin and cytochrome oxidase), and it can be detected

by an appropriate sensor (optode) and conducted to a

photomultiplier. The 2 hemoglobins have different absor-

bance characteristics, depending on their oxygenation state,

and by appropriate elaboration of the light signal at different

wavelengths, it is possible to detect the relative changes in

their concentration. This method not only gives valuable

information on the oxygenation state of the brain, but, by

measuring changes of the total hemoglobin concentration

(tHb = O2Hb + HHb), it provides an estimate of

modifications of the cerebral blood volume (CBV) [15-17].

In our department, infants undergoing surgical correction

of CDH, during the past 3 years, were treated by a standard

protocol and were systematically studied by NIRS. The aim

of this study was to analyze the possible changes in cerebral

hemodynamics and oxygenation during surgical correction

of CDH.

1. Subjects and methods

1.1. Subjects

Only infants with isolated high-risk CDH and without

associated malformations were included in this study. All

infants were managed according to a standard protocol,

based on gentle ventilation and delayed surgery, after

cardiorespiratory stabilization, which has been described

previously [3].

1.2. Neonatal stabilization

In prenatally diagnosed cases, cesarean delivery was

scheduled as close to term as possible. The neonate was

immediately intubated at birth, and gentle conventional

ventilation (PIP V25 cm H2O) (Babylog 8000 plus, Draeger,

Lubeck, Germany; or Stephanie, Ginevri, Rome, Italy) was

Table 2 Cardiorespiratory variables at the beginning (T1) and at the

T1 T2

HR (beats/min) 149.5 F 9.1 165

Sao2 (%) 94.1. F 4.6 93

Fio2 0.25 F 0.05 0.3

MABP (mm Hg) 54.7 F 7.7 55

Values are expressed as means F SD or 95% CI.a Significance of difference: P b .001.

started trying to maintain Paco2 between 40 and 60 Torr, pH

greater than 7.25, and preductal transcutaneous oxygen

saturation (tcSao2) greater than 70% in the first 2 hours of

life and between 75% and 90% in the following hours. When

PIP greater than 25 cm H2O was necessary, we shifted to

high-frequency oscillatory ventilation (3100A, Sensor Med-

ics, Viasys, Conshohocken, PA or Stephanie, Ginevri).

Inhaled nitric oxide was started only if pulmonary hyperten-

sion was present as demonstrated by right to left shunting

through the ductus arteriosus and/or tricuspid insufficiency.

Sedation and paralysis were provided with fentanyl

(2 lg/kg per hour) and pancuronium (0.l mg/kg per hour),

and surgical repair was delayed until cardiovascular and

respiratory stability had been maintained for at least 48 con-

secutive hours, without pulmonary hypertension and with

closed or very small and hemodynamically insignificant

ductus arteriosus.

1.3. Surgical intervention

The surgical procedure was performed in the neonatal

intensive care unit using an abdominal subcostal approach,

without leaving a chest tube. A diaphragmatic patch was

positioned when indicated. The ventilatory settings were

kept unchanged throughout the surgical procedure, whereas

the inspired fractional oxygen concentration (Fio2) was

adjusted, to maintain a preductal tcSao2 greater than 80%.

1.4. Cardiorespiratory monitoring

Heart rate (HR [beats per minute]), preductal tcSao2,

transcutaneous carbon dioxide tension (tcPco2 [Torr]), and

mean arterial blood pressure (MABP [mm Hg]) were

monitored and recorded continuously throughout the surgi-

cal procedure.

1.5. Cerebral monitoring

Cerebral hemodynamics and oxygenation were studied

by NIRS using a NIRO 300 monitor (Hamamatsu

Photonics Inc, Hamamatsu, Japan), consisting of an emitter

and a receiver optode of near-infrared light with an

interoptode distance of 50 mm. The light emitter carries 4

laser diodes operating at 775, 810, 847, and 919 nm, and

the detector optode has 3 photodiodes, mounted in parallel

with increasing distances, permitting the calculation of the

tissue oxygenation index (TOI) [18,19]. The optodes were

placed in a holder on the forehead and held in place by a

end (T2) of surgical repair

DT2-T1 95% CI

.2 F 14.2 15.7a 10.3-21.1

.4 F 4.4 ÿ0.7 ÿ3.1 to 1.6

7 F 0.14 0.12a 0.07-0.17

.6 F 8.1 0.9 ÿ2.7 to 4.6

Table 3 Changes in cerebral hemodynamics and oxygenation

measured by NIRS before (T1) and after (T2) surgical repair

DT2-T1 95% CI P

TOI(%) ÿ6.05 F 10.6 ÿ10.9 to ÿ1.2 b.05

O2Hb (lmol/L) ÿ10.90 F 9.7 ÿ14.0 to ÿ6.9 b.001

tHb (lmol/L) ÿ7.53 F 11.7 ÿ12.4 to ÿ2.7 b.005

HHb (lmol/L) 3.50. F 6.8 0.7-6.3 b.05

Values are expressed as mean differences between T2 and T1 with SD

and 95% CI.

Cerebral hemodynamics in congenital diaphragmatic hernia 1751

self-adhesive sticker and a headband, avoiding the temporal

muscles and the sagittal sinus; interference from ambient

light was prevented by shielding the optodes [19]. This

system allows the continuous detection of relative changes

in the cerebral concentration of tHb, O2Hb, and HHb,

(in micromoles per liter) as well as the calculation of the

TOI (TOI% = O2Hb/O2Hb + HHb) [17,20].

1.6. Data analysis

Baseline values of each variable were recorded immedi-

ately before surgery and were compared with the values

measured at the end of the surgical procedure. The 2 sets of

data were compared using the Student t test for paired

values, and the differences were considered statistically

significant for P values less than .05. Data have been

expressed as means or medians and SD or 95% confidence

intervals (95% CIs) and range where appropriate.

2. Results

2.1. Subjects

Twenty-five consecutive patients with isolated CDH

were studied: mean birth weight, 3057 F 353 g; mean

gestational age, 37.8 F 1.8; male/female newborns, 15/10;

prenatal diagnosis, 21/25; left/right side, 19/6 (Table 1).

Twenty-four infants survived to discharge.

2.2. Surgical intervention

Surgical intervention was performed at a mean age of

3.5 days (median, 2.7; range, 2-14) (Table 1). Mean duration

of the surgical procedure was 109.9 F 69.0 minutes. In

12 cases, a Dacron (Dacron Bard De Bakey, Tempe, AZ)

diaphragmatic patch was positioned. The subcostal laparot-

omy was primarily closed in all cases.

2.3. Cardiorespiratory variables

At the beginning of surgery (T1), the HR was 149.5 F

9.1 beats per minute; the tcSao2 was 94.1%F 4.6%, with an

Fio2 of 0.25 F 0.05; tcPaco2 was 49.9 F 12.8 Torr, and

MABP was 54.7F 7.7 mm Hg. At the end of surgery (T2), to

obtain similar values of tcSao2 (93.4% F 4.4%), it was

necessary to increase the Fio2 to 0.37 F 0.14 (P b .001);

HR increased significantly to 165.2 F 14.2 beats per

minute (P b .001) at T2, whereas tcPaco2 and MABP did

not change significantly (57.3 F 17.9 Torr and 55.6 F

8.1 mm Hg, respectively) (Table 2).

2.4. Cerebral monitoring

TOI, O2Hb, and tHb decreased, whereas HHb increased

significantly during surgery: DTOI = ÿ6.0% F 10.6%

(95% CI, ÿ10.9 to ÿ1.2; P b .05), DO2Hb = ÿ10.9 F

9.7 lmol/L (95%CI,ÿ14.0 toÿ6.9;P b.05),DtHb =ÿ7.5F

11.7 lmol/L (95% CI, ÿ12.4 to ÿ2.7; P b 0.001),

DHH = +3.5 F 6.8 lmol/L (95% CI, 0.71/6.3; P b .005)

(Table 3).

3. Discussion

The present study indicates that infants with isolated

severe CDH, treated with a delayed surgery and gentle

ventilation strategy, experience a moderate but statistically

significant reduction in CBV and cerebral oxygenation

during the surgical procedure.

Despite the adjustment in Fio2, to maintain a preductal

tcSao2 greater than 85%, there was a reduction in cerebral

TOI and oxygenated hemoglobin and an increase in cerebral

deoxygenated hemoglobin during the surgical procedure.

Total hemoglobin, which is an indirect index of CBV, also

decreased significantly during surgery. In particular, in all

infants, the most important modifications in cerebral

hemodynamics were observed when the herniated viscera

were repositioned into the abdomen.

The reasons for these changes are not completely clear.

One possible factor affecting cerebral circulation is Paco2

[13]: hypocapnia induces a vasoconstricting response,

whereas hypercapnia is followed by cerebral vasodilation

[21]. Our infants were mechanically ventilated since birth

with the aim to maintain their Paco2 between 40 and

60 Torr. Moreover, during surgery, the ventilatory settings

were maintained constant, trying to avoid sudden changes in

transmural pulmonary pressure and arterial blood gases.

Furthermore, the Paco2 was continuously recorded by

transcutaneous monitoring and also by arterial blood gas

analysis at the beginning and at the end of surgery, showing

only a modest and insignificant rise. Therefore, it is unlikely

that the observed changes in cerebral hemodynamics were

caused by modifications of the Paco2.

Since the most significant changes were observed during

the repositioning of the herniated viscera into the abdomen, it

is more likely that this maneuver might have interfered with

cerebral perfusion, through a compression of the inferior

vena cava and a reduction of the cardiac venous return. This

possibility is also supported by the significant increase in

HR, which could represent a compensatory phenomenon

aimed at avoiding a fall in perfusion pressure. Since the mean

arterial blood pressure did not change appreciably after

surgery, it is also possible that there was a redistribution of

the cardiac output. These factors, along with the well-known

A. Dotta et al.1752

reduced autoregulation of cerebral blood flow, typical of the

neonatal period, could have led to a decrease in cerebral

perfusion. Considering that these infants underwent surgical

correction after at least 48 hours of cardiorespiratory

stabilization, it appears that the even delayed surgery cannot

avoid moderate changes in cerebral hemodynamics in infants

with CDH. Therefore, it is reasonable to speculate that such

changes could be more marked in the presence of cardiore-

spiratory instability, particularly in preterm infants who are

less capable of autoregulating their cerebral perfusion.

The technique that we used can only measure relative

changes in cerebral circulation and oxygenation and does not

provide absolute values; furthermore, normal standards for

newborn infants are not yet available. Therefore, further and

perhaps more sophisticated studies, including appropriate

controls, are needed to clarify the clinical significance and the

possible long-term consequences of such changes in cerebral

hemodynamics. In any case, it is clear that the cerebral

circulation of infants with severe CDH seems to be

particularly sensitive to invasive maneuvers and that the

introduction of innovative therapeutic strategies for CDH

should include the evaluation of cerebral hemodynamics.

References

[1] Ivascu FA, Hirschl RB. New approaches to managing congenital

diaphragmatic hernia. Semin Perinatol 2004;28(3):185 -98.

[2] Doyle NM, Lally KP. The CDH study group and advances in the

clinical care of the patient with congenital diaphragmatic hernia.

Semin Perinatol 2004;28(3):174-84.

[3] Bagolan P, Casaccia G, Crescenzi F, et al. Impact of a current

treatment protocol on outcome of high-risk congenital diaphragmatic

hernia. J Pediatr Surg 2004;39:313 -8.

[4] Cartlidge PH, Mann NP, Kapila L. Preoperative stabilization in

congenital diaphragmatic hernia. Arch Dis Child 1986;61:1226-8.

[5] Langer JC, Filler RM, Bohn DJ, et al. Timing of surgery for

congenital diaphragmatic hernia: is emergency operation necessary?

J Pediatr Surg 1988;23:731 -4.

[6] Sakai H, Tamura M, Hosokawa Y, et al. Effect of surgical repair on

respiratory mechanics in congenital diaphragmatic hernia. J Pediatr

1987;111:432-8.

[7] Clark RH, Harding WD, Hirschl RB, et al. Current surgical

management of congenital diaphragmatic hernia: a report from the

Congenital Diaphragmatic Hernia Study Group. J Pediatr Surg 1998; 

33:1004-9.

[8] Boloker J, Batemen DA, Wung JT, et al. Congenital diaphragmatic

hernia in 120 infants treated consecutively with permissive hypercap-

nia/ spontaneous respiration/elective repair. J Pediatr Surg 2002;37: 

357 -66.

[9] Rozmiarek AJ, Qureshi FG, Cassidy L, et al. Factors influencing

survival in newborns with congenital diaphragmatic hernia: the

relative role of timing of surgery. J Pediatr Surg 2004;39:821 -4.

[10] Drummond WH, Gregory GA, Heymann MA, et al. The independent

effects of hyperventilation, tolazoline, and dopamine on infants with

persistent pulmonary hypertension. J Pediatr 1981;98:603 -11.

[11] Wung JT, Sahni R, Moffitt ST, et al. Congenital diaphragmatic hernia:

survival treated with very delayed surgery, spontaneous respiration,

and no chest tube. J Pediatr Surg 1995;30:406 -9.

[12] Langham MR, Kays DW, Beierle EA, et al. Twenty years of progress

in congenital diaphragmatic hernia at the University of Florida. Am

Surg 2003;69:45 -52.

[13] van Bel F, van de Bor M, Baan J, et al. The influence of abnormal

blood gases on cerebral blood flow velocity in the preterm newborn.

Neuropediatrics 1988;19:27-32.

[14] Jobsis FF. Non invasive, infrared monitoring of cerebral and

myocardial oxygen sufficiency and circulatory parameters. Science

1977;198:1264-74.

[15] Adcock LM, Wafelman LS, Hegmier S, et al. Neonatal intensive care

applications of near-infrared spectroscopy. Clin Perinatol 1999;26(4): 

893-902.

[16] Naulaers G, Morren G, Van Huffel S, et al. Cerebral tissue

oxygenation index in very premature infants. Arch Dis Child Fetal

Neonatal Ed 2002;87:F189-92.

[17] Weiss M, Schulz G, Fasnacht M, et al. Transcutaneously measured

near-infrared spectroscopic liver tissue oxygenation does not correlate

with hepatic venous oxygenation in children. Can J Anaesth

2002;49(8):824-9.

[18] Schulz G, Keller E, Haensse D, et al. Slow blood sampling from an

umbilical artery catheter prevents a decrease in cerebral oxygenation

in preterm newborn. Pediatrics 2003;111(1):e73-6.

[19] Suzuki S, Takasaki S, Osaki T, et al. A tissue oxygenation monitor

using NIR spatially resolved spectroscopy. Proc SPIE 1999;3597: 

582-92.

[20] Valipour A, McGown AD, Makker H, et al. Some factors affecting

cerebral tissue saturation during obstructive sleep apnoea. Eur Respir J

2002;20:444 -50.

[21] Pryds O. Control of cerebral circulation in the high-risk neonate. Ann

Neurol 1991;30:321.