review, we sought a homogenous population of patients
who underwent VV only, which is the more widely
available and commonly performed technique. Given the
wide variations in technique for VE, the availability of
VE (either due to training or practice pattern), and the
lower chance of postoperative patency and pregnancy, we
did not include VE patients in this analysis. We also
excluded studies reporting reconstruction for reasons
other than prior vasectomy.
At our institution, we perform intraoperative vasal
fl
uid analysis in the operating room. Vasal
fl
uid is
examined using an overhead camera at
±
400 magni
f
ca-
tion. We classify gross quality of vasal
fl
uid as creamy vs
cloudy vs clear. We classify microscopic examination as
whole motile sperm vs whole nonmotile sperm vs sperm
heads or short tails vs azoospermia. Based on the review of
the literature, we perform VE only in cases of intravasal
azoospermia. Gross quality of
fl
uid may be used to prog-
nosticate success with VR but has minimal in
fl
uence of
intraoperative decision making between performing a VV
or a VE. During the operation, we use a standardized form
to record intraoperative
f
ndings (
Supplementary Fig. 1
).
There were several major weaknesses in almost all the
studies analyzed. The marked heterogeneity in the de
f
-
nitions of the presence of sperm at microscopy and
patency
made
comparisons
across
studies
dif
f
cult.
Furthermore, some studies did not describe the technique
used to evaluate sperm intraoperatively. Few studies re-
ported on the quality of vasal
fl
uid and even fewer pro-
vided data on the microscopic examination of the vasal
fl
uid. Patency after VV may also depend on multiple
variables, including surgeon experience, obstructive in-
terval, age of the female partner, whether the female
partner is the same individual before and after VR, and
female factor infertility.
3,22-25
Unfortunately, there was a
lack of consistent reporting of these factors. In fact, there
was no report of female factor among studies included in
the
f
nal analysis. Because of this inconsistent reporting,
meta-analysis of other vasal
fl
uid characteristics (eg,
fl
uid
consistency) and outcomes (eg, pregnancy) was not
possible, and we were unable to statistically adjust our
primary analysis of vasal sperm to better determine
whether it provides independently meaningful informa-
tion in the operating room.
An additional weakness of our meta-analysis was that it
depended on the published literature. Because of the
relatively small number of studies available, we made the
Random Effects Model
-
4.00
0.00
2.00
4.00
6.00
8.00
Log Odds Ratio
Smith, 2014
Bolduc, 2007
Kolettis, 2005
Sigman, 2004
Sheynkin, 2000
Belker, 1991
10
468
26
51
0
437
1
83
8
1
0
49
0
36
0
6
7
50
0
19
0
0
8
33
1.95 [
-
2.33 , 6.22 ]
1.09 [
0.49 , 1.69 ]
1.14 [
-
2.86 , 5.13 ]
0.97 [
-
2.33 , 4.27 ]
0.13 [
-
3.92 , 4.17 ]
1.77 [
1.24 , 2.30 ]
1.41 [
0.84 , 1.99 ]
Author, year
+ Sperm
+P
−
P
−
Sperm
+P
−
P
Log OR
[95% CI]
Figure 2.
Association between the presence of sperm in the vasal
fl
uid during vasectomy reversal and postoperative patency
in identi
f
ed retrospective studies. The summary estimate was calculated using a random-effects model. Exponentiation of the
pooled log(OR) plotted in the
f
gure yielded an unadjusted OR of 4.1 (95% con
f
dence interval [CI], 2.3-7.3). The pooled
log(OR) calculated using a
f
xed effect model was 1.5 (95% CI, 1.1-1.9) with a corresponding OR of 4.3 (95% CI, 2.9-6.4).
Assessment of heterogeneity: Q
¼
3.4, df
¼
5,
P
¼
.6, I
2
¼
22%. CI, con
f
dence interval; OR, odds ratio; P, patent.
Figure 3.
Funnel plot assessment of publication bias. Egger
et al
13
regression analysis yielded: z
¼²
0.4;
P
¼
.7.
812
UROLOGY 85 (4), 2015