International Journal of Surgery 43 (2017) 17e25

Contents lists available at ScienceDirect

International Journal of Surgery

journal homepage: www.journal-surgery.net

Review

Comparative efficacy and safety of different circumcisions for patients

with redundant prepuce or phimosis: A network meta-analysis
Chuiguo Huang a, *, Pan Song b, Changbao Xu a, Ruofan Wang a, Lei Wei a,
Xinghua Zhao a, **
a
Department of Urology, The Second Affiliated Hospital of Zhengzhou University, Zhengzhou, China
b
Department of Urology, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China

h i g h l i g h t s

The effect of three circumcisions (CC, SRC and DCSD) on redundant prepuce or phimosis has not been conclusively studied in previous researches.

The study had collected high-quality RCTs in order to conduct an overall NMA for comparative safety and efficacy in three treatments.

The analysis shows a significant increase in the satisfaction of postoperative penile appearance after the therapy of novel circumcisions (SRC, DCSD),
comparing with CC.

The network meta-analysis confirmed that the disposable circumcision suture device could be the best choice for patients with phimosis or redundant
prepuce.

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

Article history: Background: Phimosis and redundant prepuce are defined as the inability of the foreskin to be retracted
Received 3 April 2017 behind the glans penis in uncircumcised males. To synthesize the evidence and provide the hierarchies of
Received in revised form different circumcisions for phimosis and redundant prepuce, we performed an overall network meta-
27 April 2017
analysis (NMA) based on their comparative efficacy and safety.
Accepted 27 April 2017
Material and methods: Electronic databases including PubMed, Embase, Wan Fang, VIP, CNKI and CBM
Available online 15 May 2017
database were researched from randomized controlled trials (RCTs) for redundant prepuce or phimosis.
We conducted the direct and indirect comparisons by aggregate data drug information system (ADDIS)
Keywords:
Conventional circumcision
software. Moreover, consistency models were applied to assess the differences among the male
Disposable circumcision suture device circumcision practices, and the ranks based on probabilities of intervention for the different endpoints
Shang ring were performed. Node-splitting analysis was used to test inconsistency.
Phimosis Results: Eighteen RCTs were included with 6179 participants. Compared with the conventional cir-
Redundant prepuce cumcision(CC), two new styles of circumcisions, the disposable circumcision suture device(DCSD) and
Network meta-analysis Shang Ring circumcision(SRC), provided significantly shorter operation time[DCSD: standardized mean
difference (SMD) ¼ -20.60, 95% credible interval(CI) (23.38, 17.82); SRC: SMD ¼ 19.16, 95%CI
(21.86, 16.52)], shorter wound healing time [DCSD:SMD ¼ 4.19, 95%CI (8.24,-0.04); SRC:
SMD ¼ 4.55, 95%CI (1.62, 7.57); ] and better postoperative penile appearance [DCSD: odds ratios odds
ratios (OR) ¼ 11.42, 95%CI (3.60, 37.68); SRC: OR ¼ 3.85,95%CI (1.29, 12.79)]. Additionally, DCSD showed a
lower adverse events rate than other two treatments. However, no significant difference was shown in all
surgeries for 24 h postoperative pain score. Node-splitting analysis showed that no significant incon-
sistency was existed (P > 0.05).
Conclusions: Based on the results of NMA, DCSD may be a most effective and safest choice for phimosis
and redundant prepuce. DCSD has the advantages of a shorter operation time, better postoperative penile
appearance, fewer complication and shorter wound healing time. However, with the limitations of our
study, additional multi-center RCTs are needed to evaluate the outcomes.
© 2017 Published by Elsevier Ltd on behalf of IJS Publishing Group Ltd.

* Corresponding author. Departmentof Urology, the Second Affiliated Hospital, Zhengzhou University, Jingba Road, Zhengzhou city, China.
** Corresponding author.
E-mail address: huangcg0727@163.com (C. Huang).

http://dx.doi.org/10.1016/j.ijsu.2017.04.060
1743-9191/© 2017 Published by Elsevier Ltd on behalf of IJS Publishing Group Ltd.
18 C. Huang et al. / International Journal of Surgery 43 (2017) 17e25
1. Introduction
Redundant prepuce and phimosis are common symptoms in
male adolescents who reached puberty stage, but the foreskin was
still the glans penis, all the surround and cannot be turned on. For
some men, excessive foreskin can cause inflammation and infection
on glans penis. Thereby redundant prepuce and phimosis need
removal through surgeries [1]. Dating back to more than 5000 years
ago, male circumcision (MC), represented an effective strategy for
those penile malformations, which has been performed with a
prevalence of approximately 70% in the USA and 38.7% worldwide
[1e3]. There are large volumes of published trials describing the
benefits of MC, including easier urination, improved penile topical
hygiene, increased sexual pleasure, and prevented urinary tract
infections [4,5]. Additionally, it has been demonstrated that it can
reduce sexually transmitted diseases (STDs), penile cancer and
cervical cancer associated with harboring human papilloma virus
[6,7].
There have been multiple methods of MC, such as sleeve
circumcision, dorsal slit (DS), the suture less circumcision using
tissue glue, Shang Ring circumcision (SRC) and a disposable
circumcision suture device (DCSD). One of the most common sur-
geries is the conventional circumcision (CC) which has been rec-
ommended by World Health Organization (WHO) in 2008,
including forceps guided, dorsal slit, and sleeve resection method
[8]. Even the conventional circumcision as a golden standard sur-
gery is widely performed in most MC programs. Unfortunately, it
still has such disadvantages including adverse complications,
inevitably suturing the incision, and cumbersome surgical pro-
cedure [9e11]. Moreover, those methods of the conventional
circumcision require superior surgical technique to avoid the
imperfect postoperative appearance, such as irregular hematoma
[12]. In contrast, DCSD and SRC, two novel types of disposable
circumcision devices, have substantial advantages which can
simplify surgical process, shorten operative time, reduce adverse
events, and achieve a satisfying appearance [12e16]. However, it is
still existing controversy that MC practices are more clinically
acceptable. Previous plenty researches have compared these sur-
geries in redundant prepuce and phimosis therapies, but the pre-
vious pairwise meta-analyses can not provide hierarchies of the
comparative safety and efficacy in these treatments (CC, SRC and
DCSD).
Therefore, we have collected high-quality randomized
controlled trials (RCTs) in order to conduct an overall network
meta-analysis for the comparative safety and efficacy in three
treatments. Furthermore, we provide the hierarchies of the
comparative operation time, intraoperative blood loss, 24 h post-
operative pain score, wound healing time, the incidence of adverse
events and the satisfaction on three interventions.
2. Materials and methods
2.1. Retrieval strategy
The systematic literature was performed according to the
guidelines for preferred reporting items for systematic reviews and
meta-analyses [17] (S1 file). We searched the electronic databases
including PubMed, Embase, Wan Fang database, VIP database,
Chinese National Knowledge Infrastructure (CNKI) and China
Biology Medicine (CBM) database from their inception to December
30, 2016, collecting the eligible studies for treating redundant
prepuce or phimosis without language limitation. The keywords or
MeSH search headings were used as followed: “redundant pre-
puce,” “excess foreskin,” “phimosis,” “open surgical,” “conventional
surgical,” “traditional surgical,” “disposable circumcision suture
device,” “circumcision stapler,” “DCSD,” “novel device,” “Shang
Ring,” “ring device,” “Shang huan,” “disposable anastomosis de-
vice,” and “SRC.” In addition, a manual retrieval of references from
related papers (reviews, meta-analyses and meeting reports) was
performed. And thus, all relevant articles were reviewed to
examine their eligibility. The procedure generated disagreements
which were defused through discussion with all researchers.
2.2. Selection criteria
Studies we include were in line with the following criteria: (a)
randomized controlled trials (RCTs); (b) the study included male
patients with redundant prepuce or phimosis requiring circumci-
sion; (c) studies involving the treatments among a disposable
circumcision suture device(DCSD), conventional circumcision (CC)
and Shang Ring circumcision(SRC); (d)full text available.
The following exclusion criteria were used: (a) summary, discuss
theory, letters, case reports, comments, meta-analysis, review, and
other types of research literature; (b) Duplicate publications and
data were unavailable to odds ratios (OR) or standardized mean
difference (SMD); (c) patients with genital malformations, urinary
tract infection, coagulopathy, or diabetes.
2.3. Data extraction and quality assessment
The following data was extracted and recorded in predesigned
forms from the eligible studies by two reviewers (CGH and PS)
independently: the first author's name, publication year, study
design, numbers of patients in each treatment group, ages of pa-
tients, the diagnostic criteria of patients, detail of interventions,
follow-up period, and clinical outcome measurements. The out-
comes included: (1) operation time; (2) wound healing time; (3)
intraoperative blood loss; (4) 24 h postoperative pain score; (5)
adverse event rate and (6) rate of satisfaction with postoperative
penile appearance. During the procedure, if any disagreements
have been generated, the studies would be discussed by all of us to
determine whether included or not.
The assessment tool presented by Cochrane Handbook for Sys-
tematic Reviews Interventions version 5.1.3 was applied to evaluate
the methodological quality of recruited clinical trials [18]. For
included trials, the following criteria were evaluated for risk of bias:
random sequence generation, allocation concealment, blinding of
participants and personnel, blinding of outcome assessment,
incomplete outcome data, selective reporting and other bias. Any
discrepancies from this assessment would be defused through
discussion or the third reviewer.
2.4. Data synthesis and analysis
Due to only three interventions (CC, SRC and DCSD) were
included in our analysis, we can conduct the closed triangular cir-
cular network through both direct and indirect evidences. More-
over, to calculate the consistency of direct and indirect estimates,
we performed node-splitting analysis based on ADDIS (Aggregate
Data Drug Information System, version 1.16.8) [19, 20]. The result
demonstrated that no statistical inconsistency existed in NMA
when P > 0.05.
We firstly conducted pairwise meta-analyses for studies within
DerSimonian-Laird random effects. The pair wise meta-analyses
were performed though Stata software. The pooled estimates of
odds ratios (ORs) or standardized mean difference (SMDs) and 95%
credible interval (CI) of the endpoints were shown. The Man-
teleHaenszel Chi-square based test and I [2] parameter test were
used for evaluating the heterogeneity among RCTs [20]. For these
estimates, the statistical significance should be tested using
 C. Huang et al. / International Journal of Surgery 43 (2017) 17e25
P < 0.05.
For our NMA, we used the Bayesian random-effects model to
incorporate the estimates of direct and indirect treatment com-
parisons and ranked interventions in order [21,22]. The Markov
Chains Monte Carlo (MCMC) method was performed to get the
results through ADDIS. And thus, we used the consistency model.
The models were based on 100,000 iterations for each three MCMC
chains with a burn-in period of the initial 50,000 iterations. The
relevant rank plots based on probabilities of intervention for the
different endpoints were shown by ADDIS (Fig. 4). Since the
included endpoints except the satisfaction were the directions that
lowers were better, the ranks were described that rank 1 was worst
and that rank N was best. Furthermore, the Brooks-Gelman-Rubin
method was used to evaluate convergence (Potential Scale Reduc-
tion Factors (PSRF) were limited to 1) [23].
3. Results
3.1. Eligible studies and the risks of bias assessment
A total of 758 studies were identified by searching the databases
and no additional records were identified through other sources.
After screening the duplicates, titles, abstracts and full-text, we
removed irrelevant and repeated studies, and there were 52 re-
cords appeared potentially relevant. We searched the reference lists
of all the identified studies in order to include as many relevant
articles as possible. And a total of 18 studies [13,15,16,24e38] were
eligible and included in our meta-analysis.
The PRISMA flow chart of acquisition was illustrated in Fig. 1.
Eighteen of the reports were RCTs, with a total of 1777 participants
in the DCSD group, 2589 participants in the SRC group and 1813
participants in the CC group. Baseline characteristics of the 18
studies were conducted in Table 1.
In regards to quality assessment, we make sure that all of
Fig. 1. Flow diagram of the study selection process for this meta-analysis.
19
eligible studies have described random sequence generation, and
their complete data was extracted for the primary outcome.
Nonetheless, there are only four articles reported the allocation
concealment. Moreover, only three qualified researches applied
blinding methods. The studies with imbalance in participants
among three interventions may result in other potential bias. The
assessments for bias risk were summarized quantified or qualita-
tively in Fig. 2.
3.2. The results of network meta-analysis
All the 18 studies were included in the NMA of the operation
time (Fig. 3A). The overall effect has shown that DCSD and SRC had
a statistically shorter operation time compared with CC [DCSD:
SMD ¼ 20.60, 95%CI (23.38, 17.82); SRC: SMD ¼ 19.16, 95%CI
(21.86, 16.52)]. Furthermore, the probabilities of rank plot were
as follows: CC (rank 1) was 100%; SRC (rank 2) was 86%; and DCSD
(rank 3) was 86%. Among the interventions, rank 1 is worst, and
rank 3 is best (Tables 2 and 3).
There were 16 literature reported wound healing time, recruit-
ing of 5783 patients (Fig. 3B). With CC as the reference, SMDs and
95% CI for wound healing time were as follows: SRC: 4.55(1.62,
7.57); DCSD: 4.19 (8.24,-0.04). The results demonstrated that the
patients in the CC group or DCSD group reported a markedly
shorter wound healing time than SRC group, and DCSD was supe-
rior to CC. The probabilities of CC, SRC and DCSD based on wound
healing time to be the best intervention were 3%, 0% and 97%
(Tables 2 and 3).
In total of 16 studies were analyzed for the intraoperative blood
loss (Fig. 3C). The pooled estimates revealed that DCSD group or
SRC group has the statistically less intraoperative blood loss
compared with the CC group [DCSD: SMD ¼ 9.45, 95%CI
(11.76, 7.11); SRC: SMD ¼ 9.04, 95%CI(-10.72, 7.35)]. The
probabilities of rank plot were as follows: CC (rank 1) was 100%;
 20 C. Huang et al. / International Journal of Surgery 43 (2017) 17e25

SRC (rank 2) was 65%; and DCSD (rank 3) was 65% (Tables 2 and 3).

RCT: randomized controlled trial; DCSD: disposable circumcision suture device; SRC: Shang ring circumcision; CC: conventional circumcision; NA: not available; (a)operation time; (b)intraoperative blood loss; (c)24 h
Outcome indicators

In total of 16 studies had described adverse events included

wound dehiscence, wound infection, wound edema and hema-
toma (Fig. 3D). With CC as the reference, ORs and 95% CI for the
(a)(b)(c)(d)(e)(f)

(a)(b)(c)(d)(e)(f)
(a)(b)(c)(d)(e)(f)
(a)(b)(c)(d)(e)

(a)(b)(c)(d)(e)
(a)(b)(d)(e)(f)

(a)(b)(d)(e)(f)

(a)(b)(d)(e)(f)

(a)(b)(d)(e)(f)
adverse events were as follows: SRC: 1.27(0.71, 2.31); DCSD:

(a)(b)(d)(e)
(a)(b)(d)(e)

(a)(b)(c)(d)
(a)(c)(d)(e)

(a)(b)(d)(f)
(a)(c)(d)(f)
(a)(b)(c)(f)
0.37(0.19, 0.71).A lower adverse events rate was indicated in the

(a)(d)(f)
(a)(e)(f)
DCSD group compared with other surgeries. Moreover, there was
no difference between CC and SRC[OR(95%CI):1.27(0.71,
2.31)].Furthermore, the probabilities of CC, SRC and DCSD based
30/30/30
30/30/30
30/30/30
90/90/30
Up (day)

on the participants' satisfaction for penile appearance to be the

Follow-

30/30
30/30
30/30
30/30
28/28
28/28
90/90
21/21
60/60
30/30
30/30
30/30
30/30
best intervention were 0%, 0% and 100%. However, no significant
NA

difference was detected in our overall analysis for 24 h post-

operative pain score (Fig. 3E; Tables 2 and 3).
There were a total of 13 literature reported the participants'
61/239

11/109
12/49

15/77

13/98

18/66

17/43
16/44
satisfaction for penile appearance (Fig. 3F). With CC as the refer-
7/33
Phimosis/Redundant prepuce

CC

ence, ORs and 95% CI for the satisfaction were as follows: SRC:
3.85(1.29, 12.79); DCSD: 11.42(3.60, 37.68). The pooled estimates
described that participants' satisfaction in DCSD group and SRC
15/105

70/112
40/196

11/109
25/135
16/105
75/404
10/82

31/59
9/37

group were significantly higher than CC group. Due to the satis-

SRC

faction for postoperative penile appearance was a favorable

endpoint which differed from other outcomes, ranks were
72/116
26/160

19/121
28/132

27/109
15/114

described that rank 1 is best. Therefore, the probabilities of CC,

10/39

13/98
12/80

27/63

39/81
17/43
DSCD

SRC and DCSD based on the participants' satisfaction for penile

NA

NA
NA
NA

appearance to be the best intervention were 0%, 3% and 97%

(Tables 2 and 3).
Age range

18e58

18e66
13e55

10e63

16e68

18e54
18e66

17e67
6e58

8e63
9e56

5e73

6e58

9e62
3e11

7e56

postoperative pain score; (d) wound healing time; (e)rate of satisfaction with postoperative penile appearance; (f)adverse event rate.

3.3. Consistency and convergence analysis

NA
NA

Node-splitting analysis was applied to evaluate inconsistency

26.68 ± 4.43
31.1 ± 14.8

31.7 ± 11.7
26.9 ± 10.8
25.8 ± 3.4
23.0 ± 4.3

29.3 ± 2.6
30.2 ± 5.2

by comparing the differences between direct and indirect evi-

7.1 ± 2.3

dence. After constructing the node-splitting model, we observed

that no significant inconsistency or qualitative difference was
CC

existed in this research (Table 4). That meant the consistency

27.0 ± 15.2

model was reliable. Moreover, the potential scale reduction factor

26.3 ± 2.6
20.0 ± 5.8
29.0 ± 5.1

25.0 ± 6.2
23.0 ± 5.1
32.3 ± 6.2
28.6 ± 3.2
29.6 ± 5.4
7.0 ± 2.6

(PSRF) was limited to 1, and this study can achieve good conver-
gence efficiency.
SRC

4. Discussion
27.19 ± 7.57
14.9
Age (year)

5.4
3.1
7.3
4.0
8.7
6.1
4.0
5.9

22.1 ± 2.1
23.5 ± 2.5

30.4 ± 9.1
26.2 ± 7.6

Male circumcision, a surgical removal of redundant prepuce or

±
±
±
±
±
±
±
±
±
DSCD

27.1
31.5
26.9
24.0
30.0
26.5
24.7
25.1
36.2

phimosis from the penis, is one of the oldest surgical procedures

NA

in the world. Dating back to about 2400e2300 BCE, the earliest

record of circumcision stems from Egypt, the image of the pro-
314/314/314
111/120/40

cedures of adult circumcision engraved in Ankh-Mahor's tomb at

49/46/61

92/92/92
188/182
186/236

140/120
162/158
402/322
479/354

197/201

129/120
Sample

136/84

120/60

Saqqara [1]. With the surgical techniques development, male

90/90

80/40
73/65

60/60
sizes

practices have been more and more mature. At present, conven-

tional circumcision (CC) and two novel types of disposable
circumcision devices (SRC and DCSD) were suggested to be the
most effective surgeries for redundant prepuce or phimosis
CC
CC
CC
CC
General characteristics of the 18 eligible trials involved.

[39e41]. Despite the robust observational clinical evidence,

vs
vs
vs
vs

whether the best strategy for redundant prepuce remains fairly

DCSD vs SRC
DCSD vs SRC
DCSD vs SRC
DCSD vs SRC
DCSD vs SRC
DCSD vs SRC
DCSD vs SRC
DCSD vs SRC
DCSD vs SRC
Intervention

DCSD vs CC
DCSD vs CC
DCSD vs CC
DCSD vs CC

unclear. Therefore, the relevance of treatments would need a

SRC vs CC
SRC vs CC
SRC vs CC
SRC vs CC
SRC vs CC

comprehensive analysis to comprehend.

As far as we had known, our study was performed as the first
comprehensive comparison for all reported male circumcisions.
Taken together, 18 high-quality studies containing 6179 patients
study design

RCT(2013)
RCT(2014)
RCT(2014)
RCT(2015)
RCT(2015)
RCT(2015)
RCT(2016)
RCT(2016)
RCT(2016)
RCT(2010)
RCT(2011)
RCT(2013)
RCT(2013)
RCT(2014)
RCT(2014)
RCT(2014)
RCT(2015)
RCT(2014)

met the inclusion criteria. And thus, we aimed to find the best
choice for redundant prepuce and phimosis by synthesising data,
(year)

which can balance efficacy and safety among those surgeries.

Regarding efficacy, we compared all interventions in two end-
Kanyago et al.

points (operation time and wound healing time) to evaluate

Cheng et al.
Wang et al.

Wang et al.

Wang et al.
Sokal et al.
Chen et al.
Miao et al.

Yang et al.

which surgery was the most effective method. Relative to CC,

Huo et al.
Jing et al.

Ren et al.
Pan et al.
Cao et al.

Wu et al.
Lv et al.

Li et al.

Li et al.

DCSD and SRC group have no cumbersome operation with more

Author
Table 1

steps, so that they have much shorter operation time. Further-

more, the overall effects indicated that the operation time of DCSD
Fig. 2. Risk of bias graph and summary of the included studies: (A) reviewers' judgements about each risk of bias item for eligible studies and (B) the judgements about each risk of
bias item presented as percentages across all eligible studies.

Fig. 3. Network of treatment comparisons: (A) operation time; (B) wound healing time; (C) intraoperative blood loss; (D) 24 h postoperative pain score; (E) adverse event rate and
(F)rate of satisfaction with postoperative penile appearance.
 22
C. Huang et al. / International Journal of Surgery 43 (2017) 17e25
Fig. 4. The relevant rank plots based on probabilities of intervention for the different endpoints: (A) operation time; (B) wound healing time; (C) intraoperative blood loss; (D) 24 h postoperative pain score; (E) adverse event rate and
(F)rate of satisfaction with postoperative penile appearance.
 C. Huang et al. / International Journal of Surgery 43 (2017) 17e25 23

Table 2
The network meta-analysis results for different circumcisions.

Items CC DCSD SRC

Operation time(SMD(95%Cl))
CC 1 20.60 (23.38, 17.83) 19.14 (21.86, 16.50)
DCSD 20.60 (17.83, 23.38) 1 1.45 (1.21, 4.09)
SRC 19.14 (16.50, 21.86) 1.45 (4.09, 1.21) 1

Wound healing time(SMD(95%Cl))

CC 1 4.19 (8.24,-0.04) 4.55 (1.62, 7.57)
DCSD 4.19 (0.04, 8.24) 1 8.73 (5.16, 12.23)
SRC 4.55 (7.57, 1.62) 8.73 (12.23, 5.16) 1

Intraoperative blood loss(SMD(95%Cl))

CC 1 9.45 (11.76, 7.11) 9.04 (10.72, 7.35)
DCSD 9.45 (7.11, 11.76) 1 0.40 (1.68, 2.50)
SRC 9.04 (7.35, 10.72) 0.40 (2.50, 1.68) 1

24 h postoperative pain score(SMD(95%Cl))

CC 1 2.01 (4.57, 0.60) 1.64 (3.77, 0.54)
DCSD 2.01 (0.60, 4.57) 1 0.39 (1.76, 2.51)
SRC 1.64 (0.54, 3.77) 0.39 (2.51, 1.76) 1

Adverse event rate(OR(95%Cl))

CC 1 0.37 (0.19, 0.71) 1.27 (0.71, 2.31)
DCSD 2.69 (1.41, 5.19) 1 3.43 (1.83, 6.34)
SRC 0.79 (0.43, 1.41) 0.29 (0.16, 0.55) 1

Satisfaction rate(OR(95%Cl))
CC 1 11.42 (3.60, 37.68) 3.85 (1.29, 12.79)
DCSD 0.09 (0.03, 0.28) 1 0.34 (0.11, 1.06)
SRC 0.26 (0.08, 0.78) 2.98 (0.95, 9.28) 1

Cl: 95% credible interval; SMD: standardized mean difference; OR: odds ratios.

treatment was no difference compared with SRC treatment.
We found that DCSD group was superior to CC and SRC group in
wound healing time, which may be due to the difference in the
procedures. The principle of SRC was clamping the superficial
dorsal veins and vessels between inner and outer ring-shaped
scalpel, and the foreskin can be removed by natural atrophy or
surgically, which would prolong the time for wound healing and
even longer than CC [42]. By the way, the procedure of DCSD,
imitating intestinal anastomosis, was cutting the foreskin with
annular circumcision and suturing wounds with suture staples in
the same time, which meant would reduce time [43]. In a word, our
study confirmed that DCSD can be the most effective surgery in the
aspects of the operation time and wound healing time.
In regards to safety, our analysis demonstrated that DCSD and
SRC had advantages in control bleeding. Though there was no
electric coagulation hemostasis during the operation of DCSD

Table 3
Rankings based on simulations.

Endpoints Interventions Rank 1 Rank 2 Rank 3

Operation time CC 1 0 0
(Rank 1 is worst, rank 3 is best) DCSD 0 0.13 0.87
SRC 0 0.87 0.13

Wound healing time CC 0 0.97 0.03

(Rank 1 is worst, rank 3 is best) DCSD 0 0.03 0.97
SRC 1 0 0

Intraoperative blood loss CC 1 0 0

(Rank 1 is worst, rank 3 is best) DCSD 0 0.35 0.65
SRC 0 0.65 0.35

24 h postoperative pain score CC 0.9 0.08 0.02

(Rank 1 is worst, rank 3 is best) DCSD 0.05 0.31 0.64
SRC 0.05 0.61 0.34

Adverse event rate CC 0.2 0.8 0

(Rank 1 is worst, rank 3 is best) DCSD 0 0 1
SRC 0.8 0.2 0

Satisfaction rate CC 0 0.01 0.99

(Rank 1 is best, rank 3 is worst) DCSD 0.97 0.03 0
SRC 0.03 0.96 0.01
24 C. Huang et al. / International Journal of Surgery 43 (2017) 17e25

Table 4
Results of node-splitting models.

Endpoints Comparison Direct effect Indirect effect Overall P-value

Operation time CC versus DCSD 20.94 (24.50, 17.50) 20.36 (25.83, 15.04) 20.60 (23.38, 17.82) 0.86
(SMD(95%Cl)) CC versus SRC 19.13 (22.40, 15.86) 19.64 (25.13, 14.00) 19.16 (21.86, 16.52) 0.87
DCSD versus SRC 1.26 (1.65, 4.18) 2.92 (2.08, 7.85) 1.42 (1.30, 4.10) 0.56

Wound healing time CC versus DCSD 2.68 (8.97, 3.58) 5.75 (11.57, 0.13) 4.19 (8.24,-0.04) 0.45
(SMD(95%Cl)) CC versus SRC 4.28(2.01,6.55) 5.39(1.31,8.35) 4.55(1.62,7.57) 0.64
DCSD versus SRC 8.89(6.43,11.35) 9.43(5.18,13.84) 8.73(5.16,12.23) 0.35

Intraoperative blood loss CC versus DCSD 10.69 (13.90, 7.46) 8.92 (13.01, 4.74) 9.45 (11.76, 7.11) 0.48
(SMD(95%Cl)) CC versus SRC 8.99 (10.70, 7.28) 9.95 (11.12, 5.54) 9.04 (10.72, 7.35) 0.56
DCSD versus SRC 0.15 (0.32, 0.03) 0.63 (2.93, 1.67) 0.40 (1.68, 2.50) 0.44

24 h postoperative pain score CC versus DCSD 1.22 (4.56, 2.05) 1.55 (5.60, 2.55) 2.01 (4.57, 0.60) 0.9
(SMD(95%Cl)) CC versus SRC 1.53(-4.60,1.53) 1.58(-3.79,0.59) 1.64(-3.77,0.54) 0.88
DCSD versus SRC 0.85(-0.53,2.22) 0.19(-2.40,2.75) 0.39(-1.76,2.51) 0.75

Adverse event rate CC versus DCSD 0.76 (1.51, 0.01) 1.27 (2.39, 0.13) 0.99 (1.65, 0.34) 0.45
(OR(95%Cl)) CC versus SRC 0.07 (0.55, 0.75) 0.57 (0.62, 1.82) 0.24 (0.34, 0.84) 0.45
DCSD versus SRC 1.53 (0.92, 2.14) 0.41 (0.60, 1.44) 1.23 (0.60, 1.85) 0.19

Satisfaction rate CC versus DCSD 1.87 (0.51, 3.26) 3.42 (1.34, 5.67) 2.43 (1.24, 3.68) 0.21
(OR(95%Cl)) CC versus SRC 1.72 (0.47, 3.12) 0.20 (2.00, 2.42) 1.35 (0.21, 2.56) 0.2
DCSD versus SRC 1.52 (2.59, 0.47) 0.77 (1.16, 2.83) 1.07 (2.21, 0.07) 0.14

Cl: 95% credible interval; SMD: standardized mean difference; OR: odds ratios.

treatment, the suturing staples and nails sewn the ecstatic blood
vessels up were all small, preventing bleeding loss [13]. By the way,
the outer ring in SRC treatment is clamped shut the foreskin,
covering the inner ring before the foreskin is excised and thus
hindering blood flow, so that the intraoperative blood loss can be
reduced [34]. Besides, there was no difference between DCSD and
SRC.
We also analysis the adverse events rate and postoperative pain
score to assess the safety. The result showed that DCSD had a
significantly lower rate of adverse events than other treatments,
but SRC had no difference with CC. The reason was that the patients
of SRC had needed waiting for spontaneous ring-removal, they had
to wear the device for 2e3 weeks after the operation, increasing the
risk of local complication rate, especially incision edema [42]. By
contrast, the operation of DCSD treatment was cutting the foreskin
with suturing, which meant reduce the occurrence rate of adverse
events. Furthermore, we did not find the significant difference
among three interventions in postoperative pain score, but CC with
invasive operation may be associated with increasing postoperative
pain.
Relative to clinical endpoints, patients in DCSD group were more
satisfied with the postoperative penile appearance.The results may
be as follows: (1)DCSD treatment was convenient and fast with
fewer steps; (2)the operation with suturing can effectively decrease
the risk of local complication and make the surgery more stan-
dardized. (3) as the procedure had used the more refined suturing
nails which arranged equal spacing, the wound expression can be
cosmetic appearance with even stitching density.
There were methodological strengths in our research as follow:
(1) comprehensive retrieval strategy was applied to reduce the risk
of publication bias; (2) our study was the first comparison of direct
and indirect approaches, which incorporated all available data to
evaluate the interventions more precisely; (3) the docimastic
probabilities of rank plot were utilized to distinguish the differ-
ences among all surgeries; (4) all of eligible RCTs had described
random sequence generation.
Nevertheless, our meta-analysis does have certain limitations.
Firstly, the publish data we extracted included only five types of
endpoints which had no individual patient information, so the
more detailed appraisal for variable endpoints cannot be analyzed.
Secondly, the detailed blind methods and allocation concealment
were not described in some RCTs and that may affect the validity of
overall findings. Thirdly, the heterogeneity for included outcomes
cannot be avoided. The reasons for heterogeneity as follow: (1) the
surgeons with various proficiencies had performed differences of
operator skill levels for tree surgeries; (2) objective differences
(such as pain score and satisfaction rate) existed among patients;
(3) the surgeons assessing subjectively the bleeding loss or healing
time cannot be avoided.
5. Conclusion
Our analysis found that the DCSD treatment has the advantages
of shorter operation time, no stitch removal pain, better post-
operative penile appearance, fewer complications and shorter
wound healing time than other treatments. Therefore, DCSD as the
novel disposable device might be the most efficacious and safest
choice for patients. However, as considering limitations of this
NMA, our findings need additional and high-quality studies to
validate in the future.
Conflict of interest
The authors declare no conflict of interest.
References
[1] W.D. Dunsmuir, E.M. Gordon, The history of circumcision, BJU Int. 83 (Suppl 1)
(1999) 1e12.
[2] C.P. Nelson, R. Dunn, J. Wan, J.T. Wei, The increasing incidence of newborn
circumcision: data from the nationwide inpatient sample, J. Urol. 173 (2005)
978e981.
[3] B.J. Morris, R.G. Wamai, E.B. Henebeng, et al., Estimation of country-specific
and global prevalence of male circumcision, Popul. health Metr. 14 (2016)
4e8.
[4] American Academy of Pediatrics Task Force on C, Male circumcision, Pediatrics
130 (2012) e756e85.
[5] B.J. Morris, T.E. Wiswell, Circumcision and lifetime risk of urinary tract
infection: a systematic review and meta-analysis, J. Urol. 189 (2013)
 C. Huang et al. / International Journal of Surgery 43 (2017) 17e25
2118e2124.
[6] R.C. Bailey, S. Moses, C.B. Parker, et al., Male circumcision for HIV prevention in
young men in Kisumu, Kenya: a randomised controlled trial, Lancet 369
(2007) 643e656.
[7] H.A. Weiss, S.L. Thomas, S.K. Munabi, R.J. Hayes, Male circumcision and risk of
syphilis, chancroid, and genital herpes: a systematic review and meta-anal-
ysis, Sex. Transm. Infect. 82 (2006) 101e109.
[8] H. Tim, O. Emmanuel, B. Robert, M. Palesa, O. Emmanuel, et al., Manual for
Circumcision under Local Anesthesia, Version 3.1. Ch. 5, Department of
Reproductive Health and Research, WHO, Geneva, 2009, pp. 16e31.
[9] G. Kigozi, R. Musoke, S. Watya, et al., The safety and acceptance of the PrePex
device for non-surgical adult male circumcision in Rakai, Uganda. A non-
randomized observational study, PloS one 9 (2014) e100008.
[10] A.A. Tobian, T. Adamu, J.B. Reed, et al., Voluntary medical male circumcision in
resource-constrained settings, Nat. Rev. Urol. 12 (2015) 661e670.
[11] D.R. Buwembo, R. Musoke, G. Kigozi, et al., Evaluation of the safety and effi-
ciency of the dorsal slit and sleeve methods of male circumcision provided by
physicians and clinical officers in Rakai, Uganda, BJU Int. 109 (2012) 104e108.
[12] J.H. Lei, L.R. Liu, Q. Wei, et al., Circumcision with “no-flip Shang Ring” and
“Dorsal Slit” methods for adult males: a single-centered, prospective, clinical
study, Asian J. Androl. 18 (2016) 798e802.
[13] B.D. Lv, S.G. Zhang, X.W. Zhu, et al., Disposable circumcision suture device:
clinical effect and patient satisfaction, Asian J. Androl. 16 (2014) 453e456.
[14] Y.F. Peng, Y. Cheng, G.Y. Wang, et al., Clinical application of a new device for
minimally invasive circumcision, Asian J. Androl. 10 (2008) 447e454.
[15] D.C. Sokal, P.S. Li, R. Zulu, et al., Randomized controlled trial of the shang ring
versus conventional surgical techniques for adult male circumcision: safety
and acceptability, J. Acquir. immune Defic. Syndr. 65 (2014) 447e455.
[16] J. Wang, Y. Zhou, S. Xia, et al., Safety and efficacy of a novel disposable
circumcision device: a pilot randomized controlled clinical trial at 2 centers,
Med. Sci. Monit. 20 (2014) 454e462.
[17] D. Moher, A. Liberati, J. Tetzlaff, et al., Preferred reporting items for systematic
reviews and meta-analyses: the PRISMA statement, Int. J. Surg. 8 (2010)
336e341.
[18] J.P.T. Higgins, S. Green (Eds.), Cochrane Handbook for Systematic Reviews of
Interventions Version 5.1.0, 2011. The Cochrane Collaboration; 2011. Avail-
able at: www.cochranehandbook.org. Accessed August 30, 2016.
[19] G. Valkenhoef, T. Tervonen, T. Zwinkels, B. de Brock, H. Hillege, ADDIS: a
decision support system for evidence-based medicine, Decis. Support Syst. 55
(2013) 459e475.
[20] J.P. Higgins, S.G. Thompson, J.J. Deeks, D.G. Altman, Measuring inconsistency
in meta-analyses, Bmj 327 (2003) 557e560.
[21] G. Salanti, J.P. Higgins, A.E. Ades, et al., Evaluation of networks of randomized
trials, Stat. methods Med. Res. 17 (2008) 279e301.
[22] J.P. Jansen, B. Crawford, G. Bergman, et al., Bayesian meta-analysis of multiple
treatment comparisons: an introduction to mixed treatment comparisons,
Value health J. Int. Soc. Pharmacoeconomics Outcomes Res. 11 (2008)
956e964.
[23] S.P. Brooks, A. Gelman, General methods for monitoring convergence of
iterative simulations, J. Comput. Graph. Stat. 7 (4) (1998) 434e455.
[24] Y.J. Cao, X.Z. He, G.L. Song, et al., Comparison of disposable circumcision suture
device with disposable circumcision stapler and conventional circumcision,
Chin. J. Clin. Electron. Ed. 7 (2013) 6526e6529.
[25] Z.A. Jing, Y.J. Liu, J.H. Li, et al., Randomised clinical study on comparison of the
circumcision suture device, circular stapler and traditional circumcision in the
treatment of redundant prepuce and phimosis, China J. Mod. Med. 24 (2014)
25
47e50.
[26] H.D. Miao, J.W. Lu, F.N. Lu, et al., Clinical effects of the circumcision stapler,
foreskin cerclage, and traditional circumcision: a comparative study, Natl. J.
Androl. 21 (2015) 334e337.
[27] G.F. Wu, J.J. Yan, Disposable circumcision suture device versus Shang Ring in
the treatment of phimosis or redundant prepuce, Nat. J. Urol. 21 (2015)
376e380.
[28] L.X. Wang, F.Y. Liu, L. Liu, et al., The curative effect of disposable circumcision
suture device and the Shang ring on the treatment of redundant prepuce,
Chin. J. Hum. Sex. 24 (2015) 37e41.
[29] J.L. Yang, Q.C. Chen, W. Wang, et al., Curative effect and security two types of
disposable circumcision anastomat on phimosis, Chin. J. Hum. Sex. 25 (2016)
33e35.
[30] H.C. Chen, W.C. Zhu, Q.L. Miao, et al., Comparative study on clinical effects of
disposable circumcision suture device and Shang ring for redundant prepuce
and phimosis, Chin. Gen. Prac. 14 (2016) 915e917.
[31] S.X. Wang, Z.B. Zhang, S.F. Yang, et al., Shang ring versus disposable circum-
cision suture device in the treatment of phimosis or redundant prepuce, Nat. J.
Urol. 22 (2016) 534e537.
[32] H.N. Li, J. Xu, L.M. Qu, Shang Ring circumcision versus conventional surgical
procedures: comparison of clinical effectiveness, Nat. J. Urol. 16 (2010)
325e327.
[33] Y. Cheng, Z.J. Yan, X.J. Su, et al., A clinical comparative study of Chinese Shang
Ring circumcision versus conventional circumcision, Nat. J. Urol. 32 (2011)
333e335.
[34] F. Pan, L. Pan, A. Zhang, et al., Circumcision with a novel disposable device in
Chinese children: a randomized controlled trial, Int. J. Urol. 20 (2013)
220e226.
[35] S. Kanyago, D.M. Riding, E. Mutakooha, et al., Shang Ring versus forceps-
guided adult male circumcision: a randomized, controlled effectiveness
study in southwestern Uganda, J. Acquir. immune Defic. Syndr. 64 (2013)
130e133.
[36] Y.J. Ren, X. Gao, J.N. Gong, et al., Disposable circumcision suture device of
curative effect analysis, J. Med. Theory Pract. 27 (2014) 3147e3149.
[37] S. Li, L. Zhang, D.W. Wang, et al., Clinical application of the disposable
circumcision suture device in male circumcision, Nat. J. Urol. 20 (2014)
816e819.
[38] Z.C. Huo, G. Liu, W. Wang, et al., Clinical effect of circumcision stapler in the
treatment of phimosis and redundant prepuce, Nat. J. Urol. 21 (2015)
330e333.
[39] M.R. Kaufman, M. Smelyanskaya, L.M. Van Lith, et al., Adolescent sexual and
reproductive Health services and implications for the provision of voluntary
medical male circumcision: results of a systematic literature review, PloS one
11 (2016) e0149892.
[40] D. Cao, L. Liu, Y. Hu, et al., A systematic review and meta-analysis of
circumcision with Shang Ring vs conventional circumcision, Urology 85
(2015) 799e804.
[41] Z.C. Huo, G. Liu, X.Y. Li, et al., Use of a disposable circumcision suture device
versus conventional circumcision: a systematic review and meta-analysis,
Asian J. Androl. 18 (2016) 1e6.
[42] R. Lee, E.C. Osterberg, P.S. Li, et al., Proper surgical training and grading of
complications for Shang Ring circumcision are necessary, J. Acquir. immune
Defic. Syndr. 64 (2013) e11.
[43] Z. Zhang, B. Yang, W. Yu, et al., Application of a novel disposable suture device
in circumcision: a prospective non-randomized controlled study, Int. Urol.
Nephrol. 48 (2016) 465e473.