Journal of Pediatric Surgery 57 (2022) 527–533

Contents lists available at ScienceDirect

Journal of Pediatric Surgery

journal homepage: www.elsevier.com/locate/jpedsurg.org

Delayed versus early repair of inguinal hernia in preterm infants: A

systematic review and meta-analysis
Candy SC Choo a,b, Yong Chen b,c, Merrill McHoney a,d,∗
a
University of Edinburgh, Edinburgh, United Kingdom
b
Department of Pediatric Surgery, KK Women’s and Children’s Hospital, Singapore
c
Duke-NUS Medical School, Singapore
d
Consultant Paediatric Surgeon, Royal Hospital for Sick Children Edinburgh, Edinburgh, United Kingdom

a r t i c l e i n f o a b s t r a c t

Article history: Objectives: To evaluate the clinical outcomes of herniotomy in preterm infants undergoing early versus
Received 7 October 2021 delayed repair, the risk factors for complications, and to identify best timing of surgery.
Revised 28 May 2022
Methods: Medline, Embase and Central databases were searched from inception until 25 Jan 2021 to
Accepted 4 July 2022
identify publications comparing the timing of neonatal inguinal hernia repair between early intervention
(before discharge from first hospitalization) and delayed (after first hospitalisation discharge) interven-
Keywords: tion. Inclusion criteria was preterm infants diagnosed with inguinal hernia during neonatal intensive care
Premature unit admission. Results were analyzed using fixed and random effects meta-analysis (RevManv5.4).
Inguinal hernia Results: Out of 721 articles found, six studies were included in the meta-analysis. Patients in the early
Evidence-based practice
group had lower odds of developing incarceration [odds ratio (OR) 0.43, 95% confidence interval (CI)
incarceration
0.34–0.55, I2 = 0%, p < 0.001]; but higher risk of post-operative respiratory complications (OR 4.36, 95%
complications
CI 2.13–8.94, I2 = 40%, p < 0.001). No significant differences were reported in recurrence rate (OR 3.10,
95% CI 0.90–10.64, I2 = 0%, p = 0.07) and surgical complication rate (OR 0.94, 95% CI 0.18–4.83, I2 = 0%,
p = 0.94) between early and delayed groups.
Conclusion: While early inguinal hernia repair in preterm infants reduces the risk of incarceration, it in-
creases the risk of post-operative respiratory complications compared to delayed repair. Surgeons should
discuss the risks and benefits of delaying inguinal hernia repair with the caregivers to make an informed
decision best suited to the patient physiology and circumstances.
Level of evidence: Treatment study, level 3.
© 2022 Elsevier Inc. All rights reserved.

What is currently known about this topic? • The final decision making is based on balancing the risks of
incarceration and respiratory problems, and these risks can
• Inguinal hernia repair before or after NICU discharge is de-
then be applied to individual patients and clinical settings.
bated on the balance of the risks of incarceration and surgi-
cal and anaesthetic complications
• In retrospective studies, the incidence of incarceration 1. Introduction
should be adjusted using the formula presented.
• Early inguinal hernia repair is thought to increase the risk of Inguinal hernia is a condition that commonly occurs in preterm
recurrence infants [1]. Although surgical intervention is widely accepted as the
• Both approaches are advocated by some surgeons definitive management, the optimal timing to perform surgical re-
pair remains controversial because it is a difficult to balance the
What new information is contained in this article?
risk of incarceration against the risk of intra-operative and post-
• This study quantifies the risk of incarceration and respira- operative complications in preterm infants.
tory complications in this population to help the counseling Optional operative timing is crucial in preventing incarceration.
of parents and carers. Zamakshary et al. reported a two times increase in the risk of her-
• A higher recurrence rate in the early groups approaches but nia incarceration in infants when the time of inguinal hernia repair
does not reach statistical significance. was delayed more than 14 days after diagnosis [2]. While Gholoum
et al. did not find any significant difference in the risk of incar-
∗
Corresponding address: Paediatric Surgery, John Radcliffe Hospital, Headley ceration when the inguinal hernia repair was delayed in children,
Way, Oxford OX3 9DU. and suggested that the waiting time between time of diagnosis and
E-mail address: [email protected] (M. McHoney). time of surgery may not increase the incarceration rate [3].

https://doi.org/10.1016/j.jpedsurg.2022.07.001
0022-3468/© 2022 Elsevier Inc. All rights reserved.
528 C.S. Choo, Y. Chen and M. McHoney / Journal of Pediatric Surgery 57 (2022) 527–533

Also, there is a need to balance the risk of incarceration with 2.2. Literature search
surgical and anaesthetic complications. A retrospective observa-
tional study by Moss et al. revealed low complications and recur- Studies were identified by searching the electronic databases
rence rate after surgical intervention in infants less than 2 months from Medline, Embase, Cochrane library, and reference list of ar-
old, which was comparable to those in older children [4]. However, ticles from inception to 25 Jan 2021. There was no restriction on
another retrospective observational study by Baird et al. reported the publication language. In order to improve the search findings,
that prematurity increases the risk of postoperative complications a pilot search was conducted in Medline database before applying
in infant less than 6 months old [5]. it to the rest of the electronic databases. An example of the search
The risk associated with respiratory complications after inguinal strategy can be found in appendix 1.
hernia repair have been reported by many articles. In particular,
a retrospective study by Ozdemir et al. observed 5 times increase 2.3. Study selection
in risk of postoperative apnoea in former preterm infants with a
postconceptional age (PCA) of less than 45 weeks, as compared to Two independent reviewers (C.C. and C.Y.) unblinded to the
those older than 45 weeks [6]. The authors speculated that the selected studies, screened titles and abstracts of all the articles
main predicting factor of postoperative apnoea is PCA, and rec- identified from the search results and excluded studies that were
ommended overnight observation of infants with PCA less than 45 clearly irrelevant, before reviewing the full text article of the po-
weeks. On the other hand, Vaos et al. found no significant differ- tential eligible studies. Any disagreements were resolved by con-
ence in rate of postoperative apnoea in preterm infants who had sensus with the third reviewer if needed.
inguinal hernia repair within 1 week after diagnosis as compared
to those who had surgery done more than 1 week after diagnosis
[7]. 2.4. Data extraction
One meta-analysis was published by Masoudian et al. The au-
thors pooled data from 6 retrospective studies, and concluded that Data was extracted from each included study on: country, sam-
repairing preterm infants with inguinal hernia before neonatal in- ple size, participant’s demographic (gestational age, gender, birth
tensive care unit (NICU) discharge increases the risk of recurrence weight, age at diagnosis), interventions (age at surgery, weight at
but not incarceration or surgical complication [8]. However, such surgery) and clinical outcome measures such as incarceration rate,
incidence of incarceration can only be captured in a prospective surgical complications (haematoma and wound infection), respi-
study. In retrospective data, the “incidence of incarceration” is ac- ratory complications (apnoea, respiratory insufficiency and pro-
tually the incidence of incarceration from discharge to operation in longed postoperative ventilation support), recurrence rate, reoper-
the delayed group because all incarceration before discharge would ation, spontaneous resolution and mortality. Data was extracted by
be repaired immediately and be counted as incarceration in early one reviewer and verified by a second reviewer independently.
repair [9].
Therefore, to overcome the limitations of a retrospective study, 2.5. Quality assessment
we have adjusted the incidence of incarceration in the delayed
groups and performed a meta-analysis to evaluate the clinical out- Risk of bias for included studies were evaluated using the Risk
comes of inguinal hernia repair between early surgery (before dis- Of Bias In Non-Randomized Studies – of Interventions (ROBINS-
charge from first hospitalization) and delayed surgery (after dis- I) for cohort studies [11]. Two reviewers (C.C and C.Y) assessed
charge), and to also assess the risk factors. and rated each study independently before compared the results
to reach a consensus.
2. Methods
2.6. Statistical analysis
This systematic review and meta-analysis were performed in
line with the preferred reporting items for systematic reviews The selection of included studies, characteristic of included
(PRISMA) guidelines [10]. A predefined protocol was registered studies and their risk of bias were summarised in figures and ta-
with the International Prospective Register of Systematic Reviews bles.
(PROSPERO) on 29 January 2021 (CRD42021226346). In retrospective studies, the risk of incarceration in early repair
group is likely overestimated and the incidence of incarceration
2.1. Eligibility criteria in delayed repair group underreported as shown in the supple-
mentary figure. Only prospective or randomized study can provide
Inclusion criteria were (1) article published in full-text, (2) such incidence of incarceration accurately (Appendix 2). The risk
preterm infants who were diagnosed with inguinal hernia during of incarceration between groups is also in proportion to the en-
the first hospitalisation admission and, (3) randomized controlled tire cohort of patients, but with adjustment needed. In retrospec-
trial (RCT), retrospective or observation cohort study design, and tive studies, the incidence of incarceration in early repair group is
(4) compare the outcome between early and delayed herniotomy. actually the number of incarcerations before discharge divided by
Exclusion criteria were (1) data mixed with full term infant, (2) the total number of repairs (i.e., these patients would have risked
the time of diagnosis were unclear (during first/birth hospitalisa- incarceration even if discharged and therefore relates to the en-
tion or subsequent hospital admission), and (3) outcomes were not tire group). The incidence of incarceration for delayed repair is
separated in early and delayed groups. the incidence of incarceration from discharge to operation, and be-
Primary outcomes were incarceration rate, surgical complica- cause all incarcerations before discharge would be repaired early
tions and respiratory complications. Secondary outcomes mea- and counted in the early repair an adjustment is needed. There-
sured were recurrence rate, reoperation, spontaneous resolution fore, In order to correct this error, the incidence of incarceration
and mortality. Incarcerated hernia was defined as irreducible in- in retrospective studies should be adjusted using the following for-
guinal hernia that required surgical intervention. Respiratory com- mula [9]:
plications were defined as patients with post-operative apnoea,
respiratory insufficiency or requiring prolonged postoperative ven- • Adjusted incidence of incarceration in early group = Number
tilatory support. of incarceration in early group / Total number of repairs
 C.S. Choo, Y. Chen and M. McHoney / Journal of Pediatric Surgery 57 (2022) 527–533 529

• Adjusted incidence of incarceration for delayed re- Post-operative anaesthetic complications of apnoea and respi-
pair = [Number of incarceration in early group + Number of ratory insufficiency were mentioned by 3 studies while prolonged
incarceration in delayed group + (Number of patient with- postoperative ventilation support (ventilation dependence) was re-
out incarceration in early group X Number of incarceration ported by 2 studies. The meta-analysis found that patients in the
in delayed group/Total number of repair in delayed group)] early group had higher odds of post-operative respiratory compli-
/ Total number of repairs cation as compared to the patients in the delayed group (OR 4.36,
95% CI 2.13–8.94, I2 = 40%, p < 0.001; Fig. 3). However, after ex-
Based on the above formula, the adjusted incarceration number
cluding Sulkowski et al. study for sensitivity analysis, there is no
in each group is calculated by the adjusted incident rate multiply
significant difference between early and delayed group (OR 2.44,
by number of repairs in each group.
95% CI 0.70–8.47, I2 = 30%, p < 0.16).
Pooled odds ratios (OR) were calculated for each outcome using
The surgical complications (such as haematoma and wound in-
the Mantel-Haenszel method. The confidence interval (CI) was es-
fection) reported by 2 studies revealed no significant difference
tablished at 95% and p-values of less than or equal to 0.05 were
between the early and delayed group (OR 0.94, 95% CI 0.18–4.83,
considered statistically significant. Data are presented using for-
I2 = 0%, p = 0.94).
est plots. Heterogeneity was assessed using I2 statistical and Chi
Recurrence was mentioned by 4 studies. The odds of recurrence
Square tests. An I2 value of < 25%, 26–75% and > 75% was consid-
almost achieved significant difference favoring delayed surgery
ered as low heterogeneity, moderate heterogeneity and high het-
(1.5%) as compared to early surgery group (4.7%, OR 3.10, 95% CI
erogeneity, respectively. A fixed effects model was used for low
0.90–10.64, I2 = 0%, p = 0.07; Fig. 4).
heterogeneity outcome and a random effects model was used for
Testicular atrophy rate was reported by 2 studies. However,
moderate to high heterogeneity outcome. Sensitivity analysis was
none of the patients developed it. Reoperation rate was reported
performed to exclude study with large sample size and high risk
by two studies. As both studies reported reoperation with differ-
of bias. All meta-analyzes were performed using the Cochrane Col-
ent indications, it is not possible to pool the outcome for meta-
laboration tool, RevMan5.3 [12]. Publication bias was not assessed
analysis. None of the authors reported information on spontaneous
using funnel plot as the number of studies included was <10.
resolution of inguinal hernia in patient with delayed repair. Only 1
3. Results study reported the in-hospital mortality rate and found that mor-
tality was not higher in early group. However, the p-value of 0.06
3.1. Study characteristics suggested a trend that early repair group may have higher risk of
mortality as compared to delayed repair group. Nevertheless, the
A total of 422 studies were found during the initial literature study was not able to differentiate if this was related to the hernia
search from the 3 databases. All studies identified had either the repair or other comorbidities which may be common in prematu-
title, abstract or both published in English language. After screen- rity. Hence, mortality rate was not compared in this study.
ing the titles, abstracts and removing duplicates, 24 full text arti-
cles were assessed for eligibility. Fourteen articles were excluded 3.3. Quality of evidence
as they had a different comparison group, 1 article was written as
an abstract for conference presentation with a further search re- All outcome measures, except post-operative respiratory com-
vealing no article published by the authors, and 2 articles included plications, demonstrated heterogeneity (I2 = 0%). Therefore, a fixed
participants from other age groups in their analysis. We also ex- and random effect meta-analysis were performed, respectively.
cluded an article by Crankson et al. which was included in the pre- Using the ROBINS-I tool, the overall risk of bias for the six stud-
vious meta-analysis by Masoudian et al. because some of the in- ies were found to be moderate (appendix 2). Publication bias was
guinal hernia in delayed group was diagnosed after discharge from not evaluated because of the small number of included studies
birth hospitalization [8,13]. Out of 24 articles, 6 studies fulfilled the (<10).
study eligibility criteria and were included in the study analysis
[14–19]. The study selection process is summarised in Fig. 1. 4. Discussion
All included studies were retrospective cohort study design. A
total of 2549 preterm infants were included and 308 (12%) in- Our study showed an overall incarceration rate of 12% which
fants developed incarcerated hernia. Out of all the preterm infants, is similar to the incidence reported by Olesen et al. at 11% for
84.5% (2155) were male (Table 1). The gestational age ranged from preterm children [20]. After applying a statistical formula to ad-
26.2 to 32.3 weeks and birth weight ranged from 753 g to 1750 g. just for the estimation error, our results found a 2-fold increase
The postconceptional age at surgery for the early group was be- in the risk of incarceration in preterm infant with delayed surgi-
tween 37 and 42.2 weeks, and was between 40.8 and 49 weeks in cal repair (18%) compared to the early surgical repair (9%). On the
the delayed group. Only 1 article mentioned the surgical approach other hand, the previous meta-analysis by Masoudian et al. which
(Table 2). evaluated 6 retrospective studies found that the incarceration rate
in the delayed groups is not statistically different from those with
3.2. Outcome measures early repair OR 2.15 at 95% CI 0.83–5.58, p = 0.12 [8]. This differ-
ence may be because of the lack of adjustment in the incidence of
All 6 studies reported on the incarceration rate in early and de- incarceration for retrospective studies. Hence, in Masoudian et al.
layed groups. After applying the formula for adjustment, the re- study, the incarceration rate of inguinal hernia is likely to have
sults showed that patients in early group had lower odds of de- been overestimated in the early repair group and underestimated
veloping incarceration as compared to those in the delayed group in the delayed repair group [8].
(OR 0.43, 95% CI 0.34–0.56, I2 = 0%, p < 0.001; Fig. 2). Among 6 re- Incarcerated hernia is associated with various complications in-
ports, Sulkowski et al. study has much larger sample size than all cluding ovarian, testicular and bowel ischaemia. A review of the
of the other included studies. In order to reduce the publication literature found that the risk of intestinal damage related to in-
bias, a sensitivity analysis was performed by excluding Sulkowski carcerated hernia to be between 0.1 and 12.8% in children [21–
et al. study. The result further supports that patient in early group 23]. In particular, a study by Uemura et al. showed 5% (1/21)
had lower risk of incarceration (OR 0.36, 95% CI 0.21–0.61, I2 = 0%, of the preterm infants who had inguinal hernia repair within 2
p < 0.002). weeks after the diagnosis developed incarceration, although none
530 C.S. Choo, Y. Chen and M. McHoney / Journal of Pediatric Surgery 57 (2022) 527–533

Fig. 1. PRISMA 2020 flow diagram.

Fig. 2. Forest plot comparing the odds of incarceration between early and delayed inguinal hernia repair in preterm infants.

had intestinal damage as a result [24]. However, 11% (2/19) of the by Uemura et al. was between 1993 and 1996 [24], while our cur-
preterm infants who had inguinal hernia repair later than 2 weeks rent study covers patients between 1998 and 2018. Hence, there
after diagnosis developed strangulation, and one of them had a tes- may be an improvement in the overall management of preterm in-
ticular necrosis. In our current study, a total of 308 incarcerations fant which could have reduced the risks associated with incarcer-
was reported by 6 articles, but none of them reported intestinal ation.
injury in the early or delayed repair groups. This difference may Many report that infants with PCA of less than 50 weeks have
be because of the timing of detection and management of patients higher risk of postoperative apnoea because of their greater as-
with inguinal hernia. It is also possible that this specific complica- sociation with respiratory comorbidity [14,17,25,26]. Similarly, our
tion has not been reported in all the studies. The study conducted meta-analysis revealed statistically significant differences in respi-
 C.S. Choo, Y. Chen and M. McHoney / Journal of Pediatric Surgery 57 (2022) 527–533 531

Table 1
Summary of study characteristics.

Articles Study period Country Study design Sample size (N) Gender (M/F) GA Birth weight (g) Age at diagnosis
(cGA)

Khan et al., 2009 - 2014 USA Retrospective 263 202 / 61 Early: Early: Early:37.1 ± 7.33
(42) cohort study, 28.01 ± 3.13 1070 ± 570 Delayed:43.0 ± 10
single site Delayed: Delayed:
32.3 ± 3.63 1750 ± 720
Youn et al., 1998 - 2009 South Korea Retrospective 91 72 / 18 Early: 27.5 (25 Early: 980 (430 Not reported
(40) cohort study, - 35) - 2600)
single site Delayed: 31 Delayed: 1460
(25 - 36) (740 - 3200)
Pandey et al., 2009 - 2013 USA Retrospective 39 28 / 11 Early: Early: Not reported
(43) cohort study, 26.2 ± 2.6 753 ± 158
single site Delayed: Delayed:
26.2 ± 2.7 744 ± 131
Sulkowski 1999 - 2012 USA Retrospective 2030 1760 / 270 Early: 27 (25, Early: 920 Not reported
et al., (39) cohort study, 30) (720, 1392)
Multi-site Delayed: 28 Delayed: 910
(25, 31) (709, 1343)
Takahashi 2001 - 2010 Japan Retrospective 47 28 / 19 Early: 27.6 Early: Early:36.6 ± 3 0.1
et al., (44) cohort study, 1 ± 3.1 928 ± 353 Delayed:
single site Delayed: Delayed: 38.1 ± 4.4
30.2 ± 4.0 1189 ± 486
Lee et al., (41) 2006 - 2008 USA Retrospective 80 65 / 15 Early: Early: Not reported
cohort study, 27.8 ± 3.2 1002 ± 539
single site Delayed: Delayed:
29.4 ± 3.4 1126 ± 460

Table 2
Summary of study intervention.

Articles Postconceptional Weight at Laparoscopic Length of stay Duration of

age at surgery surgery (g) repair after repair follow-up
(week) (days) (months)

Khan et al., (42) Early: 39.5 ± 4.0 Not reported Early: 19 / 115 Early:11.8 Early: 4
Delayed:40.8 ± 7.4 Delayed: 24/ (13.4) Delayed: 4.9
148 Delayed: 1.0
(2.2)
Youn et al., (40) Early: 41.9 (35.1 - Early: 2580 Not reported Not reported Not reported
56.4) (1460 - 5300)
Delayed: 43.7 (30.7 Delayed: 4300
- 84.0) (1910 - 7800)
Pandey et al., (43) Early: 41.6 ± 3.9 Not reported Not reported Not reported Not reported
Delayed:
45.4 ± 4.6
Sulkowski et al., Early:39 (37, 42) Not reported Not reported Early: 6 (2, 14) Not reported
(39) Delayed: 49 (44, Delayed: 1 (0,
55) 1)
Takahashi et al., Early: 42.2 ± 5.7 Early: Not reported Not reported Early:
(44) Delayed:48.8 ± 3.7 2919 ± 390 29.9 ± 23.6
Delayed: Delayed:
4583 ± 995 31.7 ± 27.4
Lee et al., (41) Early: 37.0 ± 6.7 Early: Not reported Early: 8 Not reported
Delayed: 2328 ± 278 Delayed: 1
44.1 ± 7.9 Delayed:
3664 ± 703

Fig. 3. Forest plot comparing the odds of respiratory complications between early and delayed inguinal hernia repair in preterm infants.
532 C.S. Choo, Y. Chen and M. McHoney / Journal of Pediatric Surgery 57 (2022) 527–533

Fig. 4. Forest plot comparing the odds of recurrence between early and delayed inguinal hernia repair in preterm infants.

ratory complications between early and delayed inguinal hernia re- One limitation of our study is that only a small number of
pair in preterm infants (p < 0.001). Preterm infants in the early re- retrospective studies were included for comparison. Furthermore,
pair group were almost 4 times more at risk of developing respira- there was inadequate information on the surgical technique used,
tory complications (47%) as compared to the preterm infant in the whether contralateral hernia exploration was done for unilateral
delayed repair group (12%). However, heterogeneity between the inguinal hernia, and the type of anaesthesia used. The informa-
studies is high. The definition of respiratory complications used by tion on the duration of follow-up and number of patients who
the identified studies were not uniform, but were grouped together were lost to follow-up is also limited. This could have significant
under respiratory complication. Post-operative apnoea and respira- effect on the outcomes of atrophy and recurrence. Moreover, the
tory insufficiency were mentioned by 3 studies [16,18,19] and pro- definition of early repair and late repair varied slightly between
longed postoperative ventilation support was the outcome of in- the articles. Three of the articles defined the timing for early re-
terest in 2 studies [14,17]. These differences in the definition for pair group as preterm infants with inguinal hernia repair prior to
respiratory complications may have led to the high heterogeneity discharge from birth hospitalisation and delayed repair group as
found. Moreover, after excluding Sulkowski et al. study with high elective inguinal hernia repair after birth hospitalisation [14,17,18].
sample size, the postoperative respiratory complications were not In comparison, the other three studies defined early repair group
significant between the groups. as inguinal hernia repair before NICU discharge and delayed repair
When we analyzed the surgical complications, we found com- as inguinal hernia repair after NICU discharge [15,16,19]. However,
parable risk between early and delayed repair groups. This may be the PCA at surgery between the articles for early and delayed re-
because of the surgeon’s expertise and/or the post-operative care pair group remained comparable. Hence, the time difference be-
provided to the preterm infants. tween the studies definition for early versus delayed inguinal her-
Although the recurrence rate in the early repair group was 3 nia repair is negligible and unlikely to affect the study outcomes.
times more than the delayed repair group, this did not reach sta- In addition, the definition for incarcerated hernia was not clearly
tistical significance (P = 0.07). This could be because of the small defined in three of the articles and they may not be classified
number of selected articles and its reported sample size. Also re- as a truly incarcerated hernia which required surgical intervention
currence rates may vary depending on the surgical technique used, [14,15,17]. The heterogeneity in the data because of these varia-
or other co-morbidities which the patient had that may have influ- tions affects interpretation, but is difficult to address with retro-
ence the risk of hernia recurrence [27,28]. Also, the recurrence rate spective studies. Lastly, the adjusted formula used in this paper to
is relatively low, and a greater sample would be needed to clarify adjust for the incarceration rate is based on statistical adjustment
the risk. and calculation. This may not reflect the true incidence of incarcer-
Only 2 out of 6 selected studies reported their number of tes- ation rate. Only prospective and randomized study design protocols
ticular atrophy and showed comparable risk of testicular atrophy can give mor accurate incarceration rate. Prospective and random-
between the early and delayed repair group [18,19]. But it remains ized studies are needed to give more accurate data.
a challenge to verify the result because of the small sample size,
lack of strong evidence from prospective RCT, inadequate duration 4.1. Differences between present study and previous meta-analysis by
of long-term follow-up and missing number of patients who were Masoudian et al
lost to follow-up [7,20,29].
Delaying hernia repair may allow a small group of patients to Currently, there is only one systematic review and meta-
develop spontaneous resolution of inguinal hernia. However, none analysis published by Masoudian et al., comparing the optimal tim-
of the studies observed and identified spontaneous resolution of ing of inguinal hernia repair in preterm infant [8]. Their study
inguinal hernia as an outcome. Thus, there is insufficient informa- found no statistical difference in the risk of incarceration between
tion to make any suggestion in this current study. While guidelines early repair group (13%) and delayed repair group (4.7%). This find-
from paediatric surgeons association have stated that inguinal her- ing differs from our present study because the authors did not
nia will not resolve spontaneously, there have been several reports make any adjustment to correct the incidence of incarceration in
supporting the possibility of spontaneous resolution in some pa- retrospective studies. The adjusted value in the present study bet-
tients [30–33]. Oudesluys-Murphy et al. found spontaneous reso- ter reflects the true incidence of incarceration rate in both early
lution of hernia in 7 out of 8 patients within 2 to 6 months and and delayed repaired groups.
reported no recurrence of inguinal hernia for a follow-up period of Another difference is that we excluded an article by Crankson
2 to 6 years [32]. A more recent article by Kurobe et al. showed et al., because the definition of delayed repair group in the article
49% (18/37) of patient with inguinal hernia spontaneously resolved consisted of former preterm infants who were diagnosed with in-
at a mean age of 6 months with 7 recurrence at a mean age of guinal hernia after discharge from first hospitalization [13]. We felt
4.8 years [33]. Therefore, delaying the surgical intervention for in- that it would be inaccurate to classify them into the delayed repair
guinal hernia may be an option to avoid surgery in some infants group because of the delay in diagnosis.
who are stable and under careful parental and surgeon supervi- They also reported an increase in the recurrence rate in early
sion. repair group of 9% compared to 2.2% in delayed repaired group
 C.S. Choo, Y. Chen and M. McHoney / Journal of Pediatric Surgery 57 (2022) 527–533 533

(p < 0.03). We found the recurrence rates at 4.7% in early surgery tional age and comorbidities. Pediatr Surg Int 2013;29:801–4. doi:10.1007/
group and 1.5% in delayed repair group, but without statistical sig- s00383- 013- 3330-8.
[7] Vaos G, Gardikis S, Kambouri K, et al. Optimal timing for repair of an in-
nificance (p = 0.07). guinal hernia in premature infants. Pediatr Surg Int 2010;26:379–85. doi:10.
Furthermore, our current study did not analyze reoperation out- 10 07/s0 0383- 010- 2573- x.
come measure because the definition for reoperation varied greatly [8] Masoudian P, Sullivan KJ, Mohamed H, et al. Optimal timing for inguinal hernia
repair in premature infants: a systematic review and meta-analysis. J Pediatr
between the studies. In the study by Takahashi et al., surgery for Surg 2019;54:1539–45. doi:10.1016/j.jpedsurg.2018.11.002.
cryptorchidism was counted as reoperation while Sulkowski et al. [9] Choo CSC, Chen Y, McHoney M. Regarding: optimal timing for inguinal hernia
included surgery for recurrence and metachronous contralateral repair in premature infants: a systematic review and meta-analysis. J Pediatr
Surg 2021;56:1083–4. doi:10.1016/j.jpedsurg.2020.12.030.
hernia as reoperation rate [14,19]. Hence, current our study did not
[10] Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an up-
pool the result together for meta-analysis. dated guideline for reporting systematic reviews. PLOS Med 2021;18:e1003583.
[11] Sterne JAC, Hernán MA, Reeves BC, et al. ROBINS-I: a tool for assessing risk of
bias in non-randomized studies of interventions. BMJ 2016;355:i4919. doi:10.
5. Conclusion
1136/bmj.i4919.
[12] The Cochrane Collaboration. Review manager 5 (RevMan 5). Version 5.3 2014.
Early inguinal hernia repair in preterm infants reduces the risk [13] Crankson S, Al Tawil K, Al Namshan M, et al. Management of inguinal hernia in
of incarceration. However, early surgery may increase the risk of premature infants: 10-year experience. J Indian Assoc Pediatr Surg 2015;20:21–
4. doi:10.4103/0971-9261.145440.
the respiratory complications as compared to delayed inguinal her- [14] Sulkowski JP, Cooper JN, Duggan EM, et al. Does timing of neonatal inguinal
nia repair. Based on current findings, delaying inguinal hernia re- hernia repair affect outcomes? J Pediatr Surg 2015;50:171–6. doi:10.1016/j.
pair may be safe for preterm infants with respiratory problems in jpedsurg.2014.10.035.
[15] Youn JK, Kim HY, Huh YJ, et al. Inguinal hernia in preterms in neonatal in-
the absence of hernia complications. Alternatively, incarceration is tensive care units: optimal timing of herniorrhaphy and necessity of con-
more common in delayed repair and patients who may have lim- tralateral exploration in unilateral presentation. J Pediatr Surg 2018;53:2155–9.
ited access to hospital on discharge may be at risk. The final de- doi:10.1016/j.jpedsurg.2018.02.056.
[16] Lee SL, Gleason JM, Sydorak RM. A critical review of premature infants with
cision making is based on balancing the risks of incarceration and inguinal hernias: optimal timing of repair, incarceration risk, and postoperative
respiratory problems, and these risks can then be applied to indi- apnea. J Pediatr Surg 2011;46:217–20. doi:10.1016/j.jpedsurg.2010.09.094.
vidual patients and clinical settings. But prospective studies, ran- [17] Khan FA, Zeidan N, Larson SD, et al. Inguinal hernias in premature neonates:
exploring optimal timing for repair. Pediatr Surg Int 2018;34:1157–61. doi:10.
domized controlled trial and long-term follow-up on the patient
10 07/s0 0383- 018- 4356- 8.
outcomes are required for better comparison. [18] Pandey R, Dako J, Venus S, et al. Early versus late inguinal hernia repair in ex-
tremely low-birthweight infants. J Matern Neonatal Med Off J Eur Assoc Peri-
nat Med Fed Asia Ocean Perinat Soc Int Soc Perinat Obstet 2017;30:2457–60.
Funding
doi:10.1080/14767058.2016.1253059.
[19] Takahashi A, Toki F, Yamamoto H, et al. Outcomes of herniotomy in prema-
This research did not receive any specific grant from funding ture infants: recent 10 year experience. Pediatr Int 2012;54:491–5. doi:10.1111/
agencies in the public, commercial, or not-for-profit sectors. j.1442-200X.2012.03607.x.
[20] Olesen CS, Mortensen LQ, Öberg S, et al. Risk of incarceration in children with
inguinal hernia: a systematic review. Hernia 2019;23:245–54.
Financial disclosure [21] Ozdamar MY, Karakus OZ. Testicular ischemia caused by incarcerated inguinal
hernia in infants: incidence, conservative treatment procedure, and follow-up.
Urol J 2017;14:4030–3. doi:10.22037/uj.v14i4.3671.
The authors have no financial relationships relevant to this ar- [22] Chang SJ, Chen JYC, Hsu CK, et al. The incidence of inguinal hernia and asso-
ticle to disclose ciated risk factors of incarceration in pediatric inguinal hernia: a nation-wide
longitudinal population-based study. Hernia 2016;20:559–63.
[23] Houben CH, Chan KWE, Mou JWC, et al. Irreducible inguinal hernia in children:
Declaration of Competing Interest how serious is it? J Pediatr Surg 2015;50:1174–6. doi:10.1016/j.jpedsurg.2014.
10.018.
The study has been submitted as a dissertation course work for [24] Uemura S, Woodward AA, Amerena R, et al. Early repair of inguinal hernia in
premature babies. Pediatr Surg Int 1999;15:36–9. doi:10.10 07/s0 03830 050507.
a master degree to the University of Edinburgh. The authors de-
[25] Cote CJ, Zaslavsky A, Downes JJ, et al. Postoperative apnea in former preterm
clare that there is no conflict of interest to disclose. infants after inguinal herniorrhaphy: a combined analysis. Anesthesiology
1995;82:809–22. doi:10.1097/0 0 0 0 0542-1995040 0 0-0 0 0 02.
[26] Walther-Larsen S, Rasmussen LS. The former preterm infant and risk of post-
Supplementary materials
operative apnoea: recommendations for management. Acta Anaesthesiol Scand
2006;50:888–93. doi:10.1111/j.1399-6576.2006.01068.x.
Supplementary material associated with this article can be [27] Taylor K, Sonderman KA, Wolf LL, et al. Hernia recurrence following in-
found, in the online version, at doi:10.1016/j.jpedsurg.2022.07.001. guinal hernia repair in children. J Pediatr Surg 2018;53:2214–18. doi:10.1016/j.
jpedsurg.2018.03.021.
[28] Eklund AS, Montgomery AK, Rasmussen IC, et al. Low Recurrence Rate After
References Laparoscopic (TEP) and Open (Lichtenstein) inguinal hernia repair: a random-
ized, multicenter trial with 5-year follow-up. Ann Surg 2009;249.
[1] Yeap E, Nataraja R, Pacilli M. Inguinal hernias in children. Aust J Gen Pract [29] de Goede B, Verhelst J, van Kempen BJ, et al. Very low birth weight is an inde-
2020;49(1–2):38–43. pendent risk factor for emergency surgery in premature infants with inguinal
[2] Zamakhshary M, To T, Guan J, et al. Risk of incarceration of inguinal hernia. J Am Coll Surg 2015;220:347–52. doi:10.1016/j.jamcollsurg.2014.11.023.
hernia among infants and young children awaiting elective surgery. CMAJ [30] CAPS. Inguinal hernias and hydroceles in infancy and childhood: a consensus
20 08;179:10 01–5. doi:10.1503/cmaj.070923. statement of the Canadian Association of paediatric surgeons. Paediatr Child
[3] Gholoum S, Baird R, Laberge JM, et al. Incarceration rates in pediatric inguinal Health 20 0 0;5:461–2. doi:10.1093/pch/5.8.461.
hernia: do not trust the coding. J Pediatr Surg 2010;45:1007–11. doi:10.1016/j. [31] Toki A, Watanabe Y, Sasaki K, et al. Adopt a wait-and-see attitude for patent
jpedsurg.2010.02.033. processus vaginalis in neonates. J Pediatr Surg 2003;38:1371–3. doi:10.1016/
[4] Moss RL, Hatch EIJ. Inguinal hernia repair in early infancy. Am J Surg s0 022-3468(03)0 0398-1.
1991;161:596–9. [32] Oudesluys-Murphy AM, Teng HT, Boxma H. Spontaneous regression of clini-
[5] Baird R, Gholoum S, Laberge JM, et al. Prematurity, not age at operation or cal inguinal hernias in preterm female infants. J Pediatr Surg 20 0 0;35:1220–1.
incarceration, impacts complication rates of inguinal hernia repair. J Pediatr doi:10.1053/jpsu.20 0 0.8757.
Surg 2011;46:908–11. doi:10.1016/j.jpedsurg.2011.02.059. [33] Kurobe M, Baba Y, Otsuka M. Inguinal hernia in very low-birthweight infants:
[6] Ozdemir T, Arıkan A. Postoperative apnea after inguinal hernia repair follow up to adolescence. Pediatr Int 2016;58:1322–7. doi:10.1111/ped.13060.
in formerly premature infants: impacts of gestational age, postconcep-