Nakagawa 2013.

Methods	Ninety‐one (91) adults between the ages of 49 and 80 years were randomised to undergo a Marcy repair (46) or Prolene Hernia System (PHS) (mesh) repair (45). Participants were evaluated and followed up postoperatively, at one week, one month, six months and then every six months up to three years postoperatively. They were assessed for recurrence, pain and infection as well as secondary endpoints including seroma/haematoma formation, testicular symptoms and objective inflammatory markers.
Participants	Adults between the ages of 49 and 80 years, with a finding of a type I‐1 or I‐2 hernia (based on the Japanese Hernia Society Classification) at the time of surgery were eligible for this study.
Interventions	Following initial dissection, the size of the hernia orifice defect was measured to confirm an I‐1 or I‐2 hernia and coexisting type II (direct) hernia was excluded. The participant was then allocated a repair group and one of the following was performed. Marcy repair The hernia sac was ligated after sufficient dissection and then the transverse fascia was sutured to tighten the inguinal ring. PHS repair The preperitoneal cavity was dissected, underlay patch inserted and spread and connector placed in internal inguinal ring. The patch was anchored to firm fascia on the anterior surface of the pubis with two sutures and to transversalis fascia, internal oblique fascia and inguinal ligament with a further six or seven sutures. All participants were given prophylactic antibiotic therapy (ampicillin) and pre‐anaesthetic buprenorphine and atropine.
Outcomes	The following outcomes were assessed postoperatively, at 1 week, 1 month, 6 months and every 6 months up to 3 years postoperatively. Hernia recurrence (defined as a palpable, reducible lump in the treated groin, with or without symptoms) Pain, numbness, difficulty walking and overall patient satisfaction were assessed by a visual analogue scale (VAS) Wound infection (diagnosed by discharge of pus from the wound) Postoperative complications: haematoma, seroma, wound swelling, testicular symptoms (up to 1 year postoperatively) Additionally, the peripheral white blood count, neutrophil count and levels of C‐reactive protein (CRP) and fibrinogen were assessed pre‐operatively and on days 3 and 7 postoperatively as an objective evaluation of inflammatory response to the surgery.
Notes	The operative time was significantly longer in PHS repair compared to Marcy repair, but no difference was seen in postoperative use of analgesics. No hernia recurrences occurred in the follow‐up period for either group. About two‐thirds of participants in both groups complained of pain at least once during the follow‐up but no differences were noted between the groups. Infeciton only occurred in one participant who had undergone a Marcy repair. Wound swelling was more common with PHS repair than Marcy repair, and occurred within one week of surgery in both groups. The mean CRP level on postop day 3 was significantly higher in the PHS group than the Marcy group, but had decreased to the same level by postoperative day 7. Two participants in the PHS repair group reported testicular symptoms but this is not expanded upon.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	At the time of surgery, participants who met the inclusion criteria and had provided written informed consent pre‐operatively were randomly allocated to a Marcy repair or PHS repair. Randomisation was performed using a computer‐generated random number table and office personnel in the Surgery Outpatient Department at Hiatsuka city were tasked with separating envelopes in the order of enrolment. Any change in the procedure after allocation was made by the attending surgeon if deemed necessary on an intention‐to‐treat basis.
Allocation concealment (selection bias)	Low risk	The attending surgeon telephoned the outpatient department from the operating theatre during surgery in order to request that an envelope be opened containing a random group allocation, and confirmed the procedure.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Although a true double‐blind study cannot be performed in the setting of surgical intervention, a single‐blinded study was achieved.
Blinding of outcome assessment (detection bias) All outcomes	High risk	No attempt was made to blind the assessors who performed the follow‐up outcome assessment; postoperative follow‐up was performed by the surgeon in charge of the study who knew the allocated surgical procedure. Primary endpoints were assessed by history and physical examination.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No participants were lost to follow‐up, however one patient in the Marcy repair group was subsequently excluded for liver cirrhosis that was recognised postoperatively and one in the PHS group due to a combined hernia being identified after allocation. Intention‐to‐treat analyses were performed.
Selective reporting (reporting bias)	Low risk	All outcomes were reported and discussed, although little detail was provided about testicular complications/symptoms.
Other bias	Low risk	No other significant biases were noted.