Optimal timing for inguinal hernia repair in premature infants: a systematic review and meta-analysis

Accepted Manuscript Optimal Timing for Inguinal Hernia Repair in Premature Infants: a Systematic Review and Meta-Analysis

Pourya Masoudian, Katrina J Sullivan, Hisham Mohamed, Ahmed Nasr PII: DOI: Reference:

S0022-3468(18)30766-8 https://doi.org/10.1016/j.jpedsurg.2018.11.002 YJPSU 58966

To appear in:

Journal of Pediatric Surgery

Received date: Revised date: Accepted date:

9 May 2018 10 October 2018 5 November 2018

Please cite this article as: Pourya Masoudian, Katrina J Sullivan, Hisham Mohamed, Ahmed Nasr , Optimal Timing for Inguinal Hernia Repair in Premature Infants: a Systematic Review and Meta-Analysis. Yjpsu (2018), https://doi.org/10.1016/ j.jpedsurg.2018.11.002

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Optimal timing for inguinal hernia repair in premature infants: A systematic review and meta-analysis

IP

T

Pourya Masoudian*a ([email protected]), Katrina J Sullivan*b ([email protected]), Hisham Mohamedb ([email protected]), Ahmed Nasra,b ([email protected])

CR

*Indicates equal contribution

US

Affiliations: aFaculty of Medicine, University of Ottawa, 451 Smyth Rd, Ottawa, ON, Canada, K1H 8M5; bDepartment of Pediatric Surgery, Children’s Hospital of Eastern Ontario, 401 Smyth Road, Ottawa, ON, Canada, K1H 8L1

AC

CE

PT

ED

M

AN

Corresponding Author: Dr. Ahmed Nasr, Children’s Hospital of Eastern Ontario, Department of Pediatric Surgery, 401 Smyth Road, Ottawa, Ontario, Canada, K1H 8L1, Tel: (613) 737-7600 x3748, Fax: (613) 738-4849, [email protected]

ACCEPTED MANUSCRIPT

US

CR

IP

T

Abstract Background: The optimal timing of repair for inguinal hernia in premature infants remains a controversial topic. Our objective was to assess the clinical effects of inguinal hernia repair done before or after neonatal intensive care unit (NICU) discharge. Methods: MEDLINE, Embase, CINAHL, and CENTRAL were searched in July 2018. Publications comparing clinical outcomes of the premature infants with inguinal hernia repair before (early) and after (delayed) NICU discharge were identified. Two reviewers independently screened studies, extracted data, and assessed for quality. Results were pooled using random effects meta-analysis. Results: Of 640 publications identified, six comparative studies assessing a total of 1761 premature infants were included. Meta-analysis indicated no statistically significant difference in incarceration rate (odds ratio (OR) 2.15, 95% confidence interval (CI) 0.835.58, I2=0%), surgical complications (OR 2.36, 95% CI 0.66-8.41, I2=0%) and other secondary complications. However, the odds of recurrence and respiratory difficulty was significantly increase in the early group compared to delayed (OR 4.12, 95% CI 1.1714.45, I2=0%; OR 3.59, 95% CI 1.10-11.75, I2=42%). Conclusions: Repair of inguinal hernia in premature infants before NICU discharge may increase the odds of recurrence, but not incarceration or surgical complications.

M

Level of Evidence: Level III

AN

Keywords: Inguinal hernia, premature, repair, timing, meta-analysis, evidence-based practice

AC

CE

PT

ED

Abbreviations: CAPS EBR: Canadian Association of Paediatric Surgeons Evidence-Based Resource CI: 95% Confidence Interval MINORS: Methodological Index for Non-Randomized Studies NICU: Neonatal Intensive Care Unit OR: Odds Ratio PCA: Post-Conceptual Age PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses

ACCEPTED MANUSCRIPT 1.0 Introduction Inguinal hernia is a common condition in premature and low-birth weight infants usually due to patent processus vaginalis, which normally closes in the last week of gestation.[13] The timing of this hernia repair in the premature population has long been an area of

T

controversy,[4] with a 2005 survey of pediatrics surgeons indicating that while 63% of

IP

surgeons prefer to repair the hernia before the hospital discharge, the other 37% would

CR

delay the surgery further.[5] Some surgeons chose to delay surgery until the infant’s postconceptual age and weight increases to reduce the risk of surgical and anesthetic

US

complications.[6-7] Additionally, there is a small chance that the hernia might

AN

spontaneously resolve over time without any surgical intervention, further supporting delayed surgery.[8] However, it has long been believed that delaying repair comes at a

M

cost of elevating the risk of incarceration, strangulation, and testicular atrophy.[9-11]

ED

We conducted a systematic review to more conclusively assess how the timing of inguinal hernia repair (before or after neonatal intensive care (NICU) discharge) affects

PT

the clinical outcomes of premature infants in order to determine the optimal timing of this

AC

2.0 Methods

CE

surgery.

The protocol for this systematic review was registered in PROSPERO on September 14, 2015 (CRD42015025902).[12] Our methods follow the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines,[13] and are described here. 2.1 Generation of research question

ACCEPTED MANUSCRIPT Pediatric surgeons were surveyed using the Delphi method to determine topics in the management inguinal hernia that required further research and/or consensus.[14] The survey was conducted on behalf of the Canadian Association of Paediatric Surgeons Evidence-Based Resource (CAPS EBR).[15] The CAPS EBR

T

(https://www.caps.ca/evidence-based-resource) is an online resource that facilitates the

CR

access up-to-date research evidence on key topics of concern.

IP

rapid uptake of good evidence into practice by providing pediatric surgeons with easy-to-

2.2 Literature search

US

An electronic search of the literature was designed by an information specialist and was

AN

conducted on July 22, 2018 in MEDLINE (1966 onwards), Embase (1980 onwards), and CENTRAL (Supplementary File 1). In addition, one reviewer (PM) hand searched

M

PubMed, Google Scholar, and references of selected articles to identify other potentially

ED

relevant studies. 2.3 Eligibility criteria

PT

We included studies published in English which compared clinical outcomes of

CE

premature infants who underwent inguinal hernia repair before NICU discharge to those who underwent hernia repair after NICU discharge. We did not include abstracts,

AC

editorials, case studies, or reviews. 2.4 Screening

Two reviewers independently screened all citations identified by the literature search for relevance. At a title and abstract level, this was done using the liberal accelerated method.[16] For assessments at the full-text level, reviewers reached consensus on articles to be included in the final analysis. Disagreements were resolved by discussion.

ACCEPTED MANUSCRIPT 2.5 Quality assessment Methodological Index for Non-Randomized Studies (MINORS)[17] was used to assess the quality of nonrandomized studies. The criterion is comprised of 12 items that evaluate the study’s validity, methods, and completeness of reporting. A comparative study is

T

assigned a score of 0-2 for each of the 12 items included, for a maximum score of 24.

IP

Higher scores are indicative of greater methodological quality. Two investigators

CR

assessed each study independently and compared their score afterwards to reach a consensus. If an agreement could not be reached, a third investigator was consulted.

US

2.6 Data extraction

AN

Data was extracted from the studies on the citation (e.g. title, year of publication), study (e.g. country/region, sample size), participants (e.g. gestational age at birth, sex),

M

intervention (e.g. age at surgery, site of hernia, interval between diagnosis and surgery),

ED

and clinical outcomes. The primary outcome of interest was incarceration rate, and secondary outcomes were surgical complications, recurrence, in-hospital mortality,

PT

testicular atrophy, respiratory difficulties (including postoperative apnea, respiratory

CE

insufficiency, and postoperative ventilation), and need for reoperation. Data was extracted by one investigator and was independently confirmed by a second.

AC

2.7 Statistical analysis

Random effects meta-analysis was used to pool post-operative events whenever possible in Review Manager 5.3.[18] The Mantel-Haenszel model was used to analyze dichotomous variables to produce an odds ratio (OR) for each outcome with a 95% confidence interval (CI). For each outcome, the heterogeneity of the study was assessed using I2 statistics. Forest plots were used to visualize the data. It was not possible to

ACCEPTED MANUSCRIPT assess publication bias using funnel plots as an insufficient number of included studies (<10) reported on the primary outcome of interest (incarceration). 2.8 Deviation from protocol It was proposed within our published protocol that we would examine the primary

T

outcome of perioperative complications (including incarceration, and recurrence) and

IP

mortality. Review of included studies indicated that no authors reported on perioperative

CR

complications, and that to combine available outcomes to result in a single “perioperative complications” outcome would be a misrepresentation of the data, requiring various

US

assumptions on the review authors part. As a result, it was concluded that the most

AN

appropriate outcomes for meta-analysis within this review would be individual outcomes,

M

rather than a combined “perioperative complications” outcome.

3.1 Study characteristics

ED

3.0 Results

PT

A total of 640 studies were identified by the search strategy and in grey literature for

CE

initial review (Figure 1). After removing duplicates and screening titles and abstracts, 65 full-text articles were assessed for eligibility. While nine studies meet a priori inclusion

AC

criteria, the authors of four of these studies had to be contacted to obtain data specific to the objective of the study. Three studies had to be excluded because the authors did not respond to emails,[19,20] or no longer had access to the data.[21] Therefore, six studies were included in the final analysis.[4, 22-26] All included studies were retrospective and comparative in design. The majority originated in the United States and were from a single centre (Table 1). A total of 1761

ACCEPTED MANUSCRIPT premature infants were evaluated within these studies, the majority of which were male with average gestational ages ranging from 26.2 to 31 weeks and average birthweights ranging from 0.74 to 1.64 kg (Table 1). Following diagnosis, patients waited an average of 2.8 to 10.7 weeks for their surgery,

T

which subsequently occurred at an average post-conceptional age range of 13 to 63

IP

weeks (Table 2). The mean weight at surgery was reported in four studies[22, 23, 25,

CR

26] and unsurprisingly displayed a lower range in the early group (2215 to 2919 g) compared to the delayed (3664 to 5810 g). The majority of hernias were located on the

US

right side or were bilateral.

AN

3.2 Risk of bias

Using the MINORS tool, all six studies were found to be of moderate quality, ranging

M

from 15 to 19 (Table 1). All six studies received a score of zero on the MINORS items of

ED

“unbiased assessment of study endpoints” and “prospective calculation of study size”. Most studies scored high on “clearly stated aim”, “inclusion of consecutive patients”,

PT

“loss to follow up less than 5%”, “adequate control group” and “contemporary groups”.

CE

Other items were reported by most studies with various degrees of adequacy (Supplementary Table 1).

AC

3.3 Primary outcome: Incarceration rate A total of five studies[22-26] reported on incarceration. Meta-analysis indicated there were no statistical difference in incarceration rate in the early group versus the delayed group (OR 2.15, 95% CI 0.83 – 5.58, I2 = 0%, p = 0.12; Figure 2). 3.4 Secondary outcomes: Surgical complications

ACCEPTED MANUSCRIPT The two studies that reported on the outcome of surgical complications[22, 25] were combined through meta-analysis, indicating no statistically significant differences between early and late surgery (OR 2.36, 95% CI 0.66 – 8.41, I2 = 0%, p = 0.19; Figure 3).

T

3.5 Secondary outcome: Recurrence rate

IP

Hernia recurrence following repair was reported in four studies[22, 24, 25, 26,]with only

CR

Youn et al.[26] providing the an average follow up time (which was 8.1 years). Metaanalysis indicated borderline statistical significance that the odds of recurrence after the

US

hernia repair is higher in the early group compared to the delayed (OR 4.12, CI 95%

AN

1.17-14.45, I=0%, p=0.03; Figure 4). 3.6 Secondary outcome: Testicular atrophy

M

Within the three studies that monitored testicular atrophy,[22, 24, 25] none of the 165

ED

premature inguinal hernia patients developed post-operative testicular atrophy. As a result, no forest plot could be generated, as there was no data to be pooled.

PT

3.7 Secondary outcome: Respiratory difficulties

CE

The odds of respiratory difficulty (which included postoperative apnea, respiratory insufficiency, and postoperative ventilation) showed borderline statistical significance

AC

favouring repair after NICU discharge compared to before (OR 3.59, 95% CI 1.10 – 11.75, I2 = 42%, p = 0.03). This was determined from the pooling of data from the five studies that reported this outcome[4, 22-25] (Figure 5). 3.8 Secondary outcome: Reoperation Pooling of Takahashi et al.[25] and Sulkowski et al.,[4] who reported on complications requiring reoperation and reoperation for hernia, respectively, indicated no statistically

ACCEPTED MANUSCRIPT significant difference between early and late surgery groups (OR 1.60, 95% CI 0.91 – 2.82, I2 = 0%, p = 0.10; Figure 6). 3.9 Secondary outcome: In-hospital mortality The in-hospital mortality rate was only reported by Sulkowski et al.[4] who measured

T

death within one year of discharge from inguinal hernia repair. The authors reported an

IP

overall very low mortality of 0.6% (6/938) in the early group and 0.2% (1/483) in the

CR

delayed; the difference was not statistically significant (p=0.27). The other five studies [22-24, 26, 27] did not comment on mortality rates intraoperatively, perioperatively, or

AN

US

post-discharge.

4.0 Discussion

M

To the best of our knowledge, this study is the first systematic review and meta-analysis

ED

conducted on timing of hernia repair in premature infants. In the past five years, there have been two literature reviews published by Wang et al.[27] and Duggan et al.[28],

PT

which did not make any conclusive recommendations regarding the optimal timing of

CE

repair.

The meta-analysis of the primary outcome, incarceration rate, indicates that there is no

AC

statistically significant difference between the rate of incarceration in neonates who undergo hernia repair before compared to after NICU discharge. This is similar to studies by Chang et al.[29] (no significant difference in incarceration rates in pediatric patients with reducible hernia with various surgical wait times) and Uemura et al.[11] (similar incarceration rate between premature infants with surgical repair ≤2 weeks of diagnosis vs. ≥2 weeks). While other studies have favoured early repair concerning incarceration

ACCEPTED MANUSCRIPT rates, their comparability is limited to our study. For example, Zamakhshary et al.[30] reported a two-fold increase in the rate of pediatric hernia incarceration when surgical delay exceeds 14 days compared to <14 days, however the patient population was not specific to premature infants. Vaos et al.[31] reported a similar finding for a wait time of

T

one week in premature infants, however only 16 patients were included in the delayed

IP

group.

CR

The overall rate of surgical complication when recurrence was included was 8%, which was reduced to 5% when recurrence was not included. Within the literature, this rate has

US

been documented as varying from 1-8%.[32-34] This meta-analysis also shows no

AN

significant difference between perioperative complications between the early and delayed group. One might expect the rate of surgical complications to be higher in the non-

M

incarcerated early group due to technical challenges in operating extremely small

ED

neonates. However, similar outcomes between early and delayed could be justified if the surgeon is experienced and comfortable with operating on small neonates. Vaos et al.[31]

PT

reported comparable rate of testicular atrophy between premature infants undergoing

CE

repair ≤1 week of diagnosis vs. >1 week. González Santacruz et al.[10] also reports similar prevalence of complications between extremely premature infants operated on ≤2

AC

weeks of diagnosis of inguinal hernia, or >2 weeks. In contrast, While Davies et al.[9] demonstrates higher incidence of testicular atrophy in infants undergoing hernia repair who are under 3 kg in weight, not all subjects were premature and there were no concurrent comparison group. Concerning other (i.e. non-surgical) complications, relatively few events were reported in early and late groups (3.4% and 3.8%, respectively), with no significant difference was identified between groups. While other

ACCEPTED MANUSCRIPT complications included fever and post-operative feeding insufficiency, respiratory distress has emerged as a major concern for these patients. Lee et al.[23] showed minimal post-op apnea for premature infants undergoing elective hernia repair; however, 13% of their cohort required prolonged (>48hr) intubation. Ultimately, the chance of such

T

complications after inguinal hernia repair in premature infants is mainly dependent on

IP

history of apnea or respiratory distress syndrome,[35-36] a variable that is not often

CR

collected in primary studies.

In our meta-analysis, we found that the early group has a borderline statistically

US

significant odds of recurrence compared to the delayed. Baird et al.[37] reported

AN

significantly higher rate of hernia recurrence after a one-year follow up in premature infants compared to term. They hypothesize that extreme friability of the hernia sac in

M

these infants along with higher number of concurrence medical issues leads to a more

ED

technically demanding repair and increases risk of post-op complications, including recurrence. In contrast, González Santacruz et al.[10] did not find any hernia recurrence

PT

in a 33-month follow-up period in either early and delayed groups. However, this rate

CE

could have potentially been due to loss to follow-up according to the authors. A borderline statistically significant difference was also found for the outcome of post-

AC

operative respiratory difficulties, where hernia patients who receive surgery prior to NICU discharge had a higher combined occurrence of apnea, respiratory insufficiency, and mechanical ventilation. A meta-analysis of eight studies and 255 patients by Cote et al.[38] showed that the chances of postoperative apnea decreased as the patient’s postconceptual age (PCA) increased. More specifically, apnea didn’t drop below 5% until PCA reached 48 weeks, and was only below 1% once PCA reached 56 weeks.

ACCEPTED MANUSCRIPT Unsurprisingly, the early group consistently had a lower PCA at the time of surgery when compared to the late group for the three studies that looked at postoperative apnea.[2224] These results may also be simply explained by more conservative ventilator management protocols by neonatologists for patients in the NICU compared to this who

T

were discharged.[23]

IP

Overall, having no significant difference in the incarceration rate and surgical

CR

complications between premature neonates having the repair before their NICU discharge compared to after is reassuring for surgeons who prefer to wait for a more suitable

US

surgical candidate. The small increase in recurrence rate found in this meta-analysis is

AN

limited by the borderline p-value of 0.03. Therefore, the evidence is not compelling that these risk of perioperative complications would decrease significantly in whom repair is

M

done after discharge. In their retrospective review, Seo et al.[21] included 96 cases of

ED

inguinal hernia in neonates with 4.2% incarceration rate and found that with active follow up by the pediatric surgeon and parental education, incarceration could be managed

PT

without morbidity allowing elective repair to be performed with more delay until the

CE

infant’s weight reach 5 kg. While this might be impractical in some centres, it is an option that could be considered at the attending surgeon’s discretion. This would require

AC

parental education and active follow-up by the surgeon or primary care provider to assess for incarceration or other complications. 4.1 Study limitations This study has some potential limitations. All included studies were retrospective, and as a result there is a potential for higher risk of bias inherent in the included studies. A RCT is currently being conducted at Vanderbilt University by Blakely et al.

ACCEPTED MANUSCRIPT (https://clinicaltrials.gov/ct2/show/NCT01678638) to determine whether early (before NICU discharge) or late (55-60 weeks post-menstrual age) inguinal hernia repair is safe in premature infants. However, the study isn’t estimated to be complete until June 2019, and no results have been posted as of yet (October 2018). Only studies published in

T

English language were included which could create language bias. Additionally, a large

IP

degree of variability current exists in defining early vs. delayed hernia repair, with studies

CR

utilizing gestational age, neonate weight, etc. in their definitions. In this systematic review we only included studies which defined early and delayed with regards to NICU

US

discharge, and some inconsistency still remained. For example, one study[4] defined

AN

early repair as before discharge from initial hospitalization rather than NICU. This allows for a small window of time in which a patient is released from NICU but is still

M

hospitalized before going home. We felt this was a minimal divergence in the definition

ED

and that it most likely would not affect the conclusion of the study. The primarily studies included in our meta-analysis also rarely provided their follow up times, and thus it is

PT

possible that certain outcomes were missed in patients due to inadequate follow up.

CE

However, this is an inherent limitation of systematic review, where authors can only use the information and data published within the primary review. Finally, publication bias

AC

could not be assessed for our primary outcome, incarceration rate, due to a small number of included studies.

5.0 Conclusion and implications This meta-analysis suggests that repair of inguinal hernia in premature infants done after NICU discharge may decrease the odds of recurrence and respiratory difficulties

ACCEPTED MANUSCRIPT compared to the repair done before the NICU discharge; however, there is no significant difference in incarceration rate and surgical complications between the two groups. This information should be considered for decision making process about timing of repair for parents and pediatric surgeons to weigh the risks and benefits of early versus delayed

T

repair. Should the decision for a delayed repair be made, parental education and close

IP

follow up by health care providers would be necessary. However, the evidence is still

CR

limited and there is a need for well-designed prospective studies in the future to

US

determine the risks and benefits associated with early and delayed timing of repair.

AN

Acknowledgements

M

We would like to thank Dr. Margaret Sampson, MLIS, PhD, AHIP, for designing our electronic search strategy

ED

Funding This research did not receive any specific grants from funding agencies in the public, commercial, or not-for-profit sectors.

AC

CE

PT

Conflict of Interest The authors do not have any conflicts of interest to declare.

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

M

AN

US

CR

IP

T

References [1] Grosfeld JL. Current concepts in inguinal hernia in infants and children. World J Surg. 1989; 13(5): 506–15 [2] Peevy KJ, Speed FA, Hoff CJ. Epidemiology of inguinal hernia in preterm neonates. Pediatrics. 1986; 77(2): 246-7. [3] Rescorla F. Hernias and umbilicus. In: Principles and practice of pediatric surgery. Philadelphia, PA: Lippincott Williams & Wilkins; 2004: 1087-96. [4] Sulkowski JP, Cooper JN, Duggan EM, Balci O, Anandalwar SP, Blakely ML, et al. Does the timing of neonatal inguinal hernia repair affect outcomes? J Pediatric Surg. 2015; 50(1): 171-6. [5] Antonoff MB, Kreykes NS, Saltzman DA, Acton RD. American Academy of Pediatrics Section on Surgery hernia survey revisited. J Pediatr Surg. 2005; 40: 1009–14. [6] Warner LO, Teitelbaum DH, Caniano DA, Vanik PE, Martino JD, Servick JD. Inguinal herniorrhaphy in young infants: Perianesthetic complications and associated preanesthetic risk factors. J Clin Anesth. 1992; 4: 455–61. [7] Cote CJ, Zaslavsky A, Downes JJ, Kurth CD, Welborn LG, Warner LO, Malviya SV. Postoperative apnea in former preterm infants after inguinal herniorrhaphy: A combined analysis. Anesthesiology. 1995; 82: 809–22. [8] Toki A, Watanabe Y, Sasaki K, Tani M, Ogura K, Wang ZC. Adopt a wait-and-see attitude for patent processus vaginalis in neonate. J Pediatr Surg. 2003; 38: 1371–3. [9] Davies BW, Fraser N, Najmaldin AS, Squire BR, Crabbe DC, Stringer MD. A prospective study of neonatal inguinal herniotomy: the problem of the postoperative hydrocele. Pediatr Surg Int. 2003; 19: 68-70. [10] González Santacruz M, Mira Navarro J, Encinas Goenechea A, Garcia Ceballos A, Sanchez Zaplana H, Jimenez Cobo B. Low prevalence of complications of delayed herniotomy in the extremely premature infant. Acta Paediatr. 2004; 93: 94-8. [11] Uemura S, Woodward AA, Amerena R, Drew J. Early repair of inguinal hernia in premature babies. Pediatr Surg Int. 1999; 15: 36-9. [12] PROSPERO: International prospective register of systematic reviews. University of York, centre for Reviews and Dissemination. 2015. Available at: http://www.crd.york.ac.uk/PROSPERO/ Accessed 20-Jun-17 [13] Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analysis: the PRISMA statement. Ann Intern Med. 2009; 151: 264-9. [14] Linstone HA, Turoff M. Delphi Method: Techniques and Applications. Boston, MA, Addison-Wesley Publishing, 1975. [15] Canadian Association of Paediatric Surgeons Evidence-Based Resource. Canadian Association of Paediatric Surgeons. 2016. Available at: https://www.caps.ca/evidence-based-resource/. Accessed 20-Jun-17 [16] Khangura S, Konnyu K, Cushman R, Grimshaw J, Moher D. Evidence summaries: The evolution of a rapid review approach. Syst Rev. 2012; 1: 10. [17] Slim K, Nini E, Forestier D, Kwiatkowski F, Panis Y, Chipponi J. Methodological index for non-randomized studies (MINORS): Development and validation of a new instrument. ANZ J Surg. 2003; 73: 712-6.

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

M

AN

US

CR

IP

T

[18] The Nordic Cochrane Centre. Review Manager (RevMan). [5.3]. 2014. The Cochrane Collaboration. Computer Program. [19] Lautz TB, Raval MV, Reynolds M. Does timing matter? A national perspective on the risk of incarceration in premature neonates with inguinal hernia. J Pediatr. 2011; 158(4): 573–7. [20] Gholoum S, Baird R, Laberge JM, Puligandla PS. Incarceration rates in pediatric inguinal hernia: do not trust the coding. J Pediatr Urol. 2010; 45: 1007-11. [21] Seo S, Takahashi T, Marusasa T, Kusafuka J, Koga H, Halibieke A, et al. Management of inguinal hernia in children can be enhanced by closer follow-up by consultant pediatric surgeons. Pediatr Surg Int. 2012; 28(1): 33-6. [22] Crankson SJ, Al Tawil K, Al Namshan M, Al Jadaan S, Baylon BJ, Gieballa M, et al. Management of inguinal hernia in premature infants: 10-year experience. J Indian Assoc Pediatr Surg. 2015; 20(1): 21–4. [23] Lee SL, Gleason JM, Sydorak RM. A critical review of premature infants with inguinal hernias: optimal timing of repair, incarceration risk, and postoperative apnea. J Pediatr Surg. 2011; 46(1): 217-20. [24] Pandey R, Dako J, Venus S, Kumar D, Mhanna M. Early versus late inguinal hernia repair in extremely low-birthweight infants. J Matern Fetal Neonatal Med. 2017; 30(20): 2457-2560f. [25] Takahashi A, Toki F, Yamamoto H, Otake S, Oki Y, Kuwano H. Outcomes of herniotomy in premature infants: recent 10 year experience. Pediatr Int. 2012; 54(4): 491-5. [26] Youn JK, Kim HY, Huh YJ, Han JW, Kim SH, Oh C, Jo AH, Park KW, Jung SE. Inguinal hernia in preterms in neonatal intensive care units: Optimal timing of herniorrhapy and necessity of contralateral exploration in unilateral presentation. J Ped Surg. 2018. E-pub ahead of print. [27] Wang KS, Committee on Fetus and Newborn, American Academy of Pediatrics, Section on Surgery, American Academy of Pediatrics. Assessment and management of inguinal hernia in infants. Pediatrics. 2012; 130(4): 768–73. [28] Duggan EM, Patel VP, Blakely ML. Inguinal hernia repair in premature infants: more questions than answers. Arch Dis Child Fetal Neonatal Ed. 2015; 100(4): F286-8. [29] Chang S-J, Chen JY-C, Hsu C-K, Chuang F-C, Yang SS-D. The incidence of inguinal hernia and associated risk factors of incarceration in pediatric inguinal hernia: a nation-wide longitudinal population-based study. Hernia. 2016; 20(4): 559-63. [30] Zamakhshary M, To T, Guan J, Langer JC. Risk of incarceration of inguinal hernia among infants and young children awaiting elective surgery. Can Med Assoc J. 2008; 179(10): 1001. [31] Vaos G, Gardikis S, Kambouri K, Sigalas I, Kourakis G, Petoussis G. Optimal timing for repair of an inguinal hernia in premature infants. Pediatr Surg Int. 2010; 26: 379-85. [32] Ein SH, Njere I, Ein A. Six thousand three hundred sixty-one pediatric inguinal hernias: a 35-year review. J Pediatr Surg. 2006; 41(5): 980–6. [33] Skinner MA, Grosfeld JL. Inguinal and umbilical hernia repair in infants and children. Surg Clin North Am. 1993; 73(3): 439–49.

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

M

AN

US

CR

IP

T

[34] Harvey MH, Johnstone MJ, Fossard DP. Inguinal herniotomy in children: a five year survey. Br J Surg. 1985; 72(6): 485–7. [35] Murphy JJ, Swanson T, Ansermino M, Milner R. The frequency of apneas in premature infants after inguinal hernia repair: do they need overnight monitoring in the intensive care unit? J Pediatr Surg. 2008; 43(5): 865–8. [36] Allen GS, Cox CS, White N, Khalil S, Rabb M, Lally KP. Postoperative respiratory complications in ex-premature infants after inguinal herniorrhaphy. J Pediatr Surg. 1998; 33(7): 1095–8. [37] Baird R, Gholoum S, Laberge J-M, Puligandla P. Prematurity, not age at operation or incarceration, impacts complication rates of inguinal hernia repair. J Ped Surg. 2011; 46(5): 908–11. [38] Coté CJ, Zadlavsky A, Downes JJ, Kurth CD, Welborn LG, Warner LO, Malviya SV. Postoperative apnea in former preterm infants after inguinal herniorrhapy. A combined anlaysis. Anesth. 1995; 82(4): 809-22.

ACCEPTED MANUSCRIPT Figure Legends Figure 1. PRISMA flow diagram for inclusion and exclusion of studies Figure 2. Forest plot comparing the odds of incarceration for early versus delayed repair of inguinal hernia in premature patients

T

Figure 3. Forest plot comparing the odds of surgical complication for early versus delayed repair of inguinal hernia in premature patients

CR

IP

Figure 4. Forest plot comparing the odds of recurrence in early and delayed repair of inguinal hernia in premature patients Figure 5. Forest plot comparing respiratory difficulties in early and delayed repair of inguinal hernia in premature patients

AC

CE

PT

ED

M

AN

US

Figure 6. Forest plot comparing re-operation in early and delayed repair of inguinal hernia in premature patients

ACCEPTED MANUSCRIPT Table 1. Summary of study characteristics and patient demographics Locatio n

Birth Weight (g)†

Gestatio n Age at Birth (weeks)†

MINOR S score*

Early: 1083±45 4 Delayed: 1433±43 6 Early: 1002±53 9 Delayed: 1126±46 0 Early: 753±158 Delayed: 744±131 Early: 875 (706.0, 1157.5) Delayed: 860 (671.0, 1110.0) Early: 928±353 Delayed: 1189±48 6 Early: 980 (4302600) Late: 1460 (7403200) Delayed: 1460 (740-

Early: 28.4±3.1 Delayed: 30.7±2.7

15

Early: 27.8±3.2 Delayed: 29.4±3.4

18

Early: 26.2±2.6 Delayed: 26.2±2.7 Early: 27 (IQR 25,29) Delayed: 27 (IQR 25, 29)

16

Early: 27.6±3.1 Delayed: 30.2±4.0

16

Early: 27.5 (2535) Late: 31 (25-36)

19

80

Pandey et United al, 2017 States [24]

Single

20092013

39

Sulkows United ki et al, States 2015 [25]

Multi

19992011

65/15

US

20062008

1421

Youn et al, 2018 [27]

Single

20012010

47

Early: 7/7 Delayed : 21/12

Single

19982009

90

Early: 13/5 Late: 59/13

AC

Takhashi Japan et al, 2012 [26]

CE

PT

ED

Early: 17/6 Delayed : 11/5 Early: 776/162 Delayed : 430/53

AN

Single

M

Lee et al, United 2011 [23] States

CR

IP

Crankson Saudi et al, Arabia 2012 [22]

Single Study Sample Sex or Perio Size of (M/F) Multid Prematur Cente e Infants r Single 1999- 84 74/10 2009

T

Study

Korea

18

ACCEPTED MANUSCRIPT 3200) Abbreviations: F: female, g: gram, IQR: Interquartile range, M: male, MINORS: Methodological Index for NonRandomized Studies, SD: Standard Deviation

AC

CE

PT

ED

M

AN

US

CR

IP

T

*Please see Supplementary Table 1 for detailed MINORS scores † Weight and Gestation Age at Birth are presented as described within their original publication and are either Mean±SD, Median (IQR), or Median (Range)

ACCEPTED MANUSCRIPT Table 2. Summary of intervention characteristics Weight at Surgery (g)†

Site of Hernia (% Rt/Lt/Bil)

Early: 2215±455 Delayed: 5810±2637

Early: 40.1/30.4/30.4 Delayed: 39.3/31.1/29.5

Lee et al, 2011 [23]

Early: 37.0±6.7 Delayed: 44.1±7.9

85% bilateral

Pandey et al, 2017 [24]

Early: 41.6±3.9 Delayed: 45.4±4.6

Early: 2328±278 Delayed: 3664±703 NR

NR

Early: 13/8.6/78.3 Delayed: 31.3/25/43.7 NR

NR

NR

Early: 7.31±4.17 Delayed: 8.6±4.4 NR

NR

NR

25.5/26.7/47.8

NR

US

NR

M

ED

Early: 2919±390 Delayed: 4583±995 Early: 2580 (1460-5300) Late: 4300 (1910-7800)

Interval between Diagnosis and Surgery (week)‡ Early: 2.8±7.4 Delayed: 9.4±76.8

AN

13 (2.7-58)

CE

Youn et al, 2018 [27]

NR

PT

Sulkowski Early: 38 (36, et al, 2015 41) [25] Delayed: 49 (43, 55) Takhashi Early: 42.2±5.7 et al, 2012 Delayed: [26] 48.8±3.7

Open

CR

Crankson Early: et al, 2012 39.5±3.05 [22] Delayed: 62.9±32.6

Open or Laparoscopic

T

Age at Surgery (week PCA)†

IP

Study

Early: 5.6±8.8 Delayed: 10.7±8.1 NR

AC

Abbreviations: Bil: Bilateral, g: Gram, IQR: Interquartile range, Lt: Left, NR: Not reported, PCA: PostConceptional Age, Rt: Right, SD: Standard Deviation † Weight and Age at Surgery are presented as described within their original publication and are either Mean±SD, Median (IQR), or Median (Range) ‡Interval between Diagnosis and Surgery is in Mean±SD

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6