Pediatr Surg Int (2002) 18: 147152

Springer-Verlag 2002

ORIGINAL ARTICLE

Vasantha H. S. Kumar Jonathan Clive

Ted S. Rosenkrantz Michael D. Bourque
Naveed Hussain

Inguinal hernia in preterm infants (32-Week Gestation)

Accepted: 17 January 2001

Abstract The current incidence of inguinal hernia (IH)

in premature infants is not well-established. It is also
unclear whether common co-morbidities in this population, i.e., chronic lung disease (CLD) or nutritional
status or both contribute to the development of IH. The
purpose of this study was to establish the epidemiologic
prole of preterm infants of 32 weeks gestational age
(GA) or less at birth with IH and determine whether the
severity of CLD or poor nutritional status predisposes
to the development of IH. Perioperative proles of
infants undergoing surgery were also reviewed. A retrospective study of 1,057 infants born at 2332 weeks GA
from January 1990 to December 1995 was done. Specic
risk and demographic factors were identied. Factors
used to determine severity of CLD were: days on intermittent mandatory ventilation (IMV); days on positive
pressure (IMV + continuous positive airway pressure);
and total number of days on supplemental oxygen.
Overall nutritional status was determined by weight gain
in g/kg per day. The incidence of IH in preterm infants
of 32 weeks GA or less who were admitted for 28 days or
more was 9.34% (65/696) prior to discharge. The incidence in infants weighing 1,500 g or less was 11.11%
(63/567) and in infants 1,000 g or less 17.39% (48/276).

V. H. S. Kumar T. S. Rosenkrantz N. Hussain (&)

Division of Neonatology, Department of Pediatrics,
University of Connecticut School of Medicine, Farmington,
CT 06030-2948, USA
J. Clive
Biostatistician, General Clinical Research Center,
University of Connecticut Health Center, Farmington,
CT 06030-3805, USA
M. D. Bourque
Division of Pediatric Surgery, Department of Surgery,
University of Connecticut School of Medicine, Farmington,
CT 06030, USA
N. Hussain
The University of Connecticut Health Center,
Division of Neonatology, Mail Code 2948,
Farmington, CT 06030-2948, USA

All parameters that determined the severity of CLD

were statistically signicant in infants with IH by univariate analysis. In a multivariate regression model,
male gender was the most important variable that was
signicantly associated with IH (odds ratio OR 9.6;
95% condence interval CI 3.9023.59), followed by
total days on supplemental oxygen (adjusted OR 1.00;
95% CI 1.011.02). Weight gain (g/kg per day) was
not signicantly dierent between the two groups. Surgical correction before discharge was well tolerated. We
conclude that the incidence of IH is GA-dependent.
Factors related to severity of CLD play a more important role than weight gain in predisposing to IH.
Keywords Inguinal hernia Chronic lung disease
Prematurity Nutrition Incidence

Introduction
It is well-recognized that inguinal hernia (IH) is a
common developmental problem in infants and children.
It is more common in low-birth-weight (LBW, < 2,500
g BW) infants than in term neonates [8, 10]. The reported incidence ranges from 5% to 30% depending on
the BW [2, 4, 15, 17]. The incidence of IH increases with
decreasing BW and is highest in extremely LBW
(ELBW) infants as reported in earlier literature [9].
Patency of the processus vaginalis is the major factor
in the development of IH [19]. By 32 weeks gestation, the
testes in males have normally entered the scrotum and
contraction of the inguinal canal around the spermatic
cord has begun [14]. Presumably, contraction of the female inguinal canal and narrowing of the canal of Nuck
occur at a similar gestational age (GA) [14]. Testicular
descent and closure of the inguinal canal are not likely to
be completed as eectively if gestation is shortened or if
growth and development of the infant are hampered.
Prematurity is the single most important predisposing
factor for the development of IH [7]. Bronchopulmonary
dysplasia (BPD) has also been associated with an

148

increased incidence of IH [11]. It has been suggested

that increased intra-abdominal pressure resulting from
chronic lung disease (CLD) or gastrointestinal dysfunction could predispose to the development of IH in
these infants [11, 16].
In preterm infants, it is not clearly known whether the
severity of CLD or the nutritional status plays a part in
the development of IHs. Preterm infants with poor
calorie/protein intake fail to thrive, leading to a lack of
tissue/muscle growth [5]. We hypothesized that severity
of CLD and nutritional status in these infants predispose to the development of IH. The purpose of this
study was to: (1) establish the epidemiologic prole of
infants with IH with respect to incidence by GA at birth
and BW; (2) analyze the impact of risk factors such as
CLD and nutritional status on the incidence of IH in
preterm infants (32 weeks gestation); and (3) analyze
the perioperative morbidity in such infants who underwent hernia repair.

Materials and methods

In a retrospective review of the hospital course of 1,057 preterm
infants with GAs between 23 and 32 completed weeks at birth, data
were retrieved by a combination of chart and database review. The
study period was from 1 January 1990 to 31 December 1995; 696
infants remained in the unit for 28 days or more. All infants were
managed at a single tertiary-care facility, the John Dempsey Hospital at The University of Connecticut Health Center, Farmington,
CT. This is a high-risk referral center serving the populations of
north-central and northeastern Connecticut.
The demographic prole of the infants studied included analyses of the incidence of IH by GA at birth, BW, race, and gender
during their stay in the nursery. For comparison, the infants with
IH were matched with infants of similar BW, GA, and gender who
did not have a hernia. The severity of CLD was assessed by the
following parameters: days on intermittent mandatory ventilation
(IMV); days on positive pressure (PP) = days on IMV + conFig. 1 Incidence of inguinal
hernia in all infants 2332 weeks
gestation admitted for 28 days
19901995. Overall incidence
9.34% (65/696), decrease with
increasing gestational age (linear
regression analysis)

tinuous positive airway pressure (CPAP); days on nasal cannula

air-ow support or nasal cannula supplemental oxygen support,
and total days on supplemental oxygen. The nutritional status of
these infants was assessed by weight gain (discharge weight admission weight) during the hospital stay in days and expressed as
g/kg BW per day.
A group of four pediatric surgeons were involved in the preoperative evaluation, operation, postoperative monitoring, and
follow-up of these infants. Timing of surgery was based on the
readiness of the infant for discharge and not on a particular weight
or post-conceptional age. The discharge criteria included: (1)
physiological stability of vital parameters including temperature
control; (2) ability to feed by mouth and gain weight on a consistent basis; and (3) absence of apnea/bradycardia spells for at least 7
days. General anesthesia was induced and maintained with isourane or sevourane inhalation in infants who were intubated. In
non-intubated infants, spinal or epidural anesthesia was given with
hyperbaric tetracaine, sometimes combined with intravenous ketamine sedation. Immediate postoperative pain relief was provided
by local inltration or caudal infusion of bupivacaine.
From the total admissions (n 1,057), the incidence of IH was
calculated based on number of infants still admitted at 28 days of
life or more (n 696). Selected study parameters were compared
between the two groups of infants (IH and controls) matched for
BW, gender, and GA at birth. Statistical analyses were done using
the chi-square and unpaired t-test to analyze individual variables
and outcome between various groups. Results were expressed as
mean standard deviation and a P value of less than 0.05 was
considered signicant. Variables that were found to be signicant
by univariate analysis were included in the regression model for
multivariate analysis. Unadjusted odds ratios and adjusted odds
ratios (Adj. OR) using 95% condence intervals (CI) were determined by stepwise multiple logistic regression analysis.

Results
Incidence of IH by (GA) at birth
The overall incidence of IH in infants 32 weeks GA or
less, who were still in the unit at over 28 days of life was
9.34% (65/696). Figure 1 shows the incidence of IH in

149

each GA group. The incidence was highest in the

youngest group and decreased with increasing GA, and
this was signicant by linear regression analysis (P<
0.0001; R2 0.895). The dierences in gender were
obvious, with males constituting 87.70% (57/65) and
females 12.30% (8/65) of all IH cases. The pre-discharge
incidence in males was 15.40% (57/370) and in females
was 2.45% (8/326).

Incidence of inguinal hernia by BW

We included all infants 1,500 g or less who were admitted for 28 days or more for calculating the incidence
of IH by BW. The incidence in very LBW (VLBW)
(1,500 g) infants was 11.11% (63/567) and in ELBW
(1,000 g) 17.39% (48/276). Figure 2 shows the incidence of IH by dierent BW categories. The incidence
was highest in the sub-group 7511,000 g and lowest in
the sub-group >1,500 g. The relationship between BW
and incidence of IH was signicant by linear regression
analysis (P < 0.03; R2 0.717). Of the infants with
hernias, 9.23% (6/65) were small for gestational age
(SGA). All SGA infants with hernias were males and 28
weeks GA or more.

Unilateral versus bilateral IH

Approximately 50% (33/65) of cases presented as bilateral IH prior to surgery. The rest were unilateral, with
31.80% (21/65) right-sided and 18.20% (12/65) leftsided.

Fig. 2 Incidence of inguinal

hernia in infants admitted for
28 days by BW categories
19901995. Overall incidence for
1,500 g group 11.11% (63/567),
decrease with increasing BW
(linear regression analysis)

Ethnicity and incidence

The incidence of IH was similar among all races and
ethnic groups (data not shown).
Impact of respiratory problems and nutritional status
We analyzed the impact of CLD and nutritional status
on infants with and without IH. Table 1 shows the
demographic prole of all infants with and without IH
matched for BW, GA, and gender. By univariate analysis, there were signicant dierences in days on positive
pressure (PP), days on supplemental oxygen, total days
on oxygen support, and length of hospital stay between
the two groups. There were no signicant dierences in
weight gain between the groups. By regression analysis,
infants with IH had been exposed to more days on
oxygen (OR 1.01; 95% CI 1.001.02) compared to
infants without IH.
Surgery for IH
Surgery prior to discharge was performed in 81.5% (53/
65) of all infants with IH. Of the remaining patients 2/65
infants (3%) died from causes unrelated to the IH, while
in the neonatal intensive care unit (NICU). Surgery was
scheduled after discharge from the NICU in 10/65
infants (15.5%). Of the infants who underwent surgery
in the hospital, only 1 had developed an incarceration
and needed surgery on an emergency basis at 3 weeks of
age. The rest underwent elective surgery 35 days prior
to discharge. In infants with a unilateral IH, the con-

150
Table 1 Study parameters in
Hernia and non-hernia groups
(NS not signicant, values
mean standard deviation,
P signicant at <0.05)

Variable

Hernia group
(n = 65)

Non-hernia group
(n = 65)

P value

Gestational age (weeks)

Birth weight (g)
Gender (male)
Days on mechanical ventilation
Days on positive pressure
Days on supplemental oxygen
Total days on oxygen support
Length of stay (days)
Weight gain (g/kg/day)

26.66
943.27
57
26.98
41.97
60.12
102.09
113.54
9.93

26.60
930.98
54
20.97
32.22
44.68
76.87
91.43
10.45

NS a
NS a
NS b
NS a
< 0.02 a
< 0.02 a
< 0.01 a
< 0.003 a
NS a

a
b

2.17
272.42
(81.81%)
17.40
21.11
31.98
47.15
37.47
2.90

Statistical analysis by unpaired t-test

Statistical analysis by chi-square test

tralateral side was explored in the majority of cases

(>90%). While bilateral IH was diagnosed in 50% of
infants prior to surgery, it was found in 80% of infants
at surgery, and all these infants underwent bilateral
inguinal herniorrhaphy. The mean weight at surgery was
3,033 968 g.
Anesthesia
Of the infants who underwent surgery, 32% (17/53) were
given general anesthesia and 68% (36/53) were not intubated and surgery was done under spinal or epidural
anesthesia. Of the latter infants, none developed apnea/
bradycardia episodes postoperatively. Of the 17 infants
who were intubated and received general anesthesia, 15
were extubated by 25 h and none of these developed
apnea/bradycardia spells postoperatively. Of the remaining 2, 1 with severe CLD remained intubated for 48
h and developed apnea/bradycardia episodes for 48 h
post-extubation; the other, who had undergone emergency surgery for incarceration of the hernia at 3 weeks
of life, was intubated for 8 days after surgery secondary
to underlying lung disease not related to anesthesia. No
other complications related to anesthesia were noted.
Recovery from surgery was uncomplicated to hospital
discharge in all cases.

Discussion
The incidence of IH in full-term infants ranges from 1%
to 5% [10]. The estimated incidence in preterm infants
has ranged from 8% to 30% and has been found to vary
with BW, GA, and length of follow-up [2, 4, 15, 17].
Table 2 Incidence of inguinal
hernia in infants 1,500 g BW
literature review

2.41
346.58
(87.70%)
23.29
30.79
45.86
68.51
45.61
2.09

These previous studies examined infants of greater BWs

and GAs. This study has the largest number of patients
and it extends the information to the lowest surviving
GA groups by reporting the incidence of IH by GA in
infants born between 23 and 32 weeks' gestation. We
believe that reporting the incidence by GA is important,
as the development and maturation of the inguinal canal
and processus vaginalis is related to GA rather than BW.
The incidence of IH was highest at 23 weeks GA, and
gradually decreased with increasing GA. From our observation, infants of 28 weeks GA or less were at the
highest risk for IH. Infants with 2932 weeks GA were
at lower risk, and the risk increased with lower BW,
SGA status, and male gender.
Most studies have reported the incidence of IH by
BW. Table 2 shows the incidence in infants of 1,500 g or
less reported in the literature compared to our study.
Our study compares favorably with the other reports
except that of Boocock and Todd [2], which was a retrospective survey of surgical registries and thus did not
account for infants who had not yet undergone surgery.
In ELBW infants (<1,000 g at birth), Harper et al.
(19671972; n 37) reported the pre-discharge incidence of IH as 30% [9]. In comparison, the incidence in
ELBW infants in our study was signicantly lower at
17.39%. Apart from the fact that the actual incidence
was lower in our study, it is also notable that the trends
in decreasing incidence with increasing BW in our study
also correspond closely to those described by Peevy
et al. [15]. Despite the fact that our study population
included the most preterm infants, the reasons why our
report documents a lower incidence may be twofold:
recent advances in neonatal care have decreased respiratory and nutritional morbidity; and our study reports
the incidence only during primary hospitalization in the

Author

Follow-up of infants

Incidence
in males

Incidence
in females

Incidence in
all infants

Rajput et al. (19771987);

n = 1,391
Boocock et al. (1981); n = 172
Darlow et al. (19831985); n = 95
Our data (19901995); n = 567

20 months corrected
age
6 months of age
13 years after birth
Primary hospitalization

26%

7%

16%

14%
NA
18.56%

2%
NA
2.9%

8.72%
18.90%
11.11%

151

nursery, and does not take into account the fact that
LBW infants, particularly boys, are at risk of developing
IH not only in infancy, but also up to 8 years of age and
beyond [12].
Among the risk factors studied, increased intra-abdominal pressure (IAP) is probably the most important
factor that predisposes to the development of IH
[16, 1820]. The fact that infants are predominantly
abdominal breathers makes it more likely that in the
presence of CLD, which increases IAP, an IH is more
likely to occur. The concentration and duration of
supplemental oxygen is a very good indicator of the
severity of CLD [1]. Of the measures of CLD in our
study, by regression model analysis only total oxygen
days were signicantly increased in infants with IH. The
length of hospital stay was signicantly longer in the
hernia group, which may in part be explained by the fact
that they had more severe lung disease (more days on
supplemental oxygen) and that they had to stay additional days for surgery and post-operative recovery.
Yeo and Gray reported that infants with respiratory
distress syndrome and those requiring mechanical ventilation were signicantly predisposed to the development of IH, especially males [27]. In another study,
bronchopulmonary dysplasia was associated with a
higher incidence of IH [11]. We have shown here that
factors associated with CLD such as days on mechanical
ventilation, CPAP, and supplemental oxygen were signicantly greater in infants who developed IH. We
therefore conclude that the severity of CLD plays an
important part in the development of IH in premature
infants.
In a study by Powell et al. [16], the best predictor
model gave a risk of IH that was inversely related to
BW, gender (M:F 11.5:1), intravenous feeding
(Y:N 3.1:1), and absence of respiratory distress
(Y:N 2.4:1), implying that the predominant predisposing factors for IH were nutritional rather than
respiratory, which is contrary to our ndings. We
studied weight gain expressed as g/kg-BW per day and
did not nd any dierence in weight gain in infants who
did or did not develop IHs. Although other factors
including feeding intolerance, gastroesophageal reux,
necrotizing enterocolotis, sepsis, total parental nutrition,
or caloric intake were not taken into account in this
study, we could still determine that ELBW infants with
IH did no worse than infants without IH with respect to
weight gain. We speculate that poor weight gain per se
may not increase the risk of IH, but factors that
contribute to failure to thrive (e.g., CLD, sepsis) may be
associated with IH by some other mechanism.
For infants in our study, surgery was performed
within 35 days of anticipated discharge from the unit
except when it was in response to a complication of the
hernia (incarceration, which occurred in only 1 infant).
By the time of surgery, all infants were on full enteric
nutrition and most of them were breathing room air.
Even though all of them required supplemental oxygen
during surgery, their postoperative oxygen require-

ments within 24 h of surgery were no dierent than

their preoperative requirements. Most infants with a
unilateral hernia had contralateral groin exploration
followed by bilateral inguinal herniorraphy when
needed. Although unilateral exploration has been advocated by some surgeons based on the theoretical increase in the risk of infertility by intervention on both
sides [13], many surgeons currently practice routine
contralateral exploration in infants and young children
due to the high incidence of bilateral IH at follow-up
[26]. This was validated in our study, where pre-operatively approximately 50% were diagnosed as having
bilateral IH, while, on exploration over 80% were
found to be bilateral. Other studies have shown that
failure to examine the contralateral inguinal ring can
result in subsequent development of a clinically
apparent contralateral IH in approximately 10% or
more of children undergoing unilateral repair. The risk
increases to 40% if repair is done before 1 year of age
[2022, 26].
Preterm infants are at risk for developing apnea and
bradycardia following general anesthesia. The risk is
greater in those with a history of apnea, but decreases
with post-conceptional age [3, 6, 23, 25]. The only
patient who developed apnea and bradycardia in our
study was a former 23 week premature infant who was
38 weeks post-conceptional age at surgery. He
remained intubated for 48 h postoperatively. The
observation that apnea/bradycardia spells do not constitute a major problem in infants undergoing repair
under spinal/epidural anesthesia [24] was conrmed in
our study: none of the infants who had spinal/epidural
anesthesia developed postoperative apnea or bradycardia. The relatively high body weight of infants at the
time of surgery may also have contributed to the
relatively low incidence of anesthetic and postoperative
morbidity.
We conclude that the incidence of IH is inversely
related to the BW and GA at birth. The pre-discharge
incidence of IH has decreased in ELBW infants (<1,000
g BW) over the past 2 decades. Factors related to
severity of CLD (e.g., days on PP support, days on
supplemental oxygen) play a more important role than
weight gain in increasing the risk for IH. IHs can safely
be corrected by surgery using spinal anesthesia prior to
discharge from the hospital.
Acknowledgements We would like to thank Marlene Holman,
R.N., Paula Gendreau, R.N., and Barbara Westman, Administrative Program Assistant, for the scrupulous maintenance of our
neonatal database.

References
1. Bhutani VK, Abbasi S (1992) Long-term pulmonary consequences in survivors with bronchopulmonary dysplasia. Clin
Perinatol 19: 649671
2. Boocock GR, Todd PJ (1985) Inguinal IHs are common in
preterm infants. Arch Dis Child 60: 669670

152
3. Cote CJ, Zaslavsky A, Downes JJ, Kurth CD, Welborn LG,
Warner LO, Malviya SV (1995) Postoperative apnea in former
preterm infants after inguinal herniorrhaphy. A combined
analysis. Anesthesiology 82: 809822
4. Darlow BA, Dawson KP, Mogridge N (1987) Inguinal hernia
and low birthweight. N Z Med J 100: 492494
5. Georgie MK, Amarnath UM, Mills MM (1989) Determinants
of arm muscle and fat accretion during the rst postnatal
month in preterm newborn infants. J Pediatr Gastroenterol
Nutr 9: 219224
6. Gollin G, Bell C, Dubose R, Touloukian RJ, Seashore JH,
Hughes CW, Oh TH, Fleming J, O'Connor T (1993) Predictors
of postoperative respiratory complications in premature infants
after inguinal herniorrhaphy. J Pediatr Surg 28: 244247
7. Grosfeld JL (1989) Current concepts in inguinal hernia in
infants and children. World J Surg 13: 506515
8. Grosfeld JL (1989) Groin IH in infants and children. In: Nyhus
LM, Condon RE (eds) Hernia, Lippincott, Philadelphia, pp
8196
9. Harper RG, Garcia A, Sia C (1975) Inguinal hernia: a common
problem of premature infants weighing 1,000 grams or less at
birth. Pediatrics 56: 112115
10. Holder TM, Ashcraft KW (1980) Groin hernias and hydroceles. In: Textbook of Pediatric Surgery, Saunders, Philadelphia,
pp 594608
11. Hrabovsky EE, Sigmund WR, Mullett MK (1983) Inguinal
Hernia in the stressed premature infant. Perinatol Neonatol 7:
4551
12. Kitchen WH, Doyle LW, Ford GW (1991) Inguinal hernia in
very low birthweight children: a continuing risk to age 8 years.
J Paediatr Child Health 27: 300301
13. Matsuda T, Muguruma K, Hiura Y, Okuno H, Shichiri Y,
Yoshida O (1998) Seminal tract obstruction caused by childhood inguinal herniorrhaphy: results of microsurgical reanastomosis. J Urol 159: 837840
14. Moore KL (1977) Development of the inguinal canals. In: The
developing human. Saunders, Philadelphia, pp 239240,
253254

15. Peevy KJ, Speed FA, Ho CJ (1986) Epidemiology of inguinal

hernia in preterm neonates. Pediatrics 77: 246247
16. Powell TG, Hallows JA, Cooke RW, Pharoah PO (1986) Why
do so many small infants develop an inguinal hernia? Arch Dis
Child 61: 991995
17. Rajput A, Gauderer MW, Hack M (1992) Inguinal hernias in
very low BW infants: incidence and timing of repair. J Pediatr
Surg 27: 13221324
18. Rescorla FJ, Grosfeld JL (1984) Inguinal hernia repair in the
perinatal period and early infancy: clinical considerations.
J Pediatr Surg 19: 832837
19. Rowe MI, Copelson LW, Clatworthy HW (1969) The patent
processus vaginalis and the inguinal hernia. J Pediatr Surg 4:
102107
20. Rowe MI, Marchildon MB (1981) Inguinal hernia and hydrocele in infants and children. Surg Clin North Am 61:
11371145
21. Sparkman RS (1962) Bilateral exploration in inguinal hernia in
juvenile patients. Surgery 51: 393406
22. Steinau G, Schleef J, Lambertz M, Schumpelick V (1997) Incidence of contralateral inguinal hernias in infancy and childhood. Langenbecks Arch Chir 382: 252256
23. Steward DJ (1982) Preterm infants are more prone to complications following minor surgery than are term infants. Anesthesiology 56: 304306
24. Tiryaki T, Baskin D, Bulut M (1998) Operative complications of hernia repair in childhood. Pediatr Surg Int 13:
160161
25. Welborn LG, Rice LJ, Hannallah RS, Broadman LM, Ruttimann UE, Fink R (1990) Postoperative apnea in former preterm infants: prospective comparison of spinal and general
anesthesia. Anesthesiology 72: 838842
26. Wiener ES, Touloukian RJ, Rodgers BM, Grosfeld JL, Smith
EI, Ziegler MM, Coran AG (1996) Hernia survey of the Section
on Surgery of the American Academy of Pediatrics. J Pediatr
Surg 31: 11661169
27. Yeo CL, Gray PH (1994) Inguinal hernia in extremely preterm
infants. J Paediatr Child Health 30: 412413

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.