Laparoscopic ovarian drilling as first line of treatment
in infertile women with PCOS.
Cleemann L, Lauszus FF, Trolle B
Dept. of Gynaecology and Obstetrics
Holstebro Hospital, Denmark
Short title: Laparoscopic ovarian drilling and PCOS
Keywords: infertility; polycystic ovarian syndrome; laparoscopic ovarian drilling; pregnancy rate.
Address for correspondence:
Finn Lauszus, M.D.
Dept. of Gynaecology and Obstetrics
Holstebro Hospital, Laegaardsvej 12
7500 Holstebro, Denmark
Phone: +45 99125247
FAX: +45 99125200
Laparoscopic ovarian drilling (LOD) is used as first line of treatment, second line of treatment after
patients have proven resistance to clomiphene or third line of treatment after failed ovulation
induction with gonadotropins. We present the postoperative pregnancy rates of 57 women to
evaluate a potential optimal time of LOD together with the other treatment regimens of infertile
women with polycystic ovarian syndrome (PCOS). Data on the preoperative and operative
treatment, background data was evaluated for influencing pregnancy rates.
The pregnancy rate was 61 % among women with PCOS who had LOD. No
difference was found in the clinical data between the women who became pregnant and the ones
who did not. Likewise no difference was found between the women who had preoperative treatment
and the ones who had none. The median time to pregnancy after LOD was 135 days. Forty-nine
percent (28/57) of the women had been treated medically with or without insemination prior to
surgery. Thirty-nine percent of these women became pregnant after the operation without any
further infertility treatment. Forty-three percent (12/28) were clomiphene resistant before operation
and forty-two percent of these women became pregnant after LOD.
LOD alone resolves the infertility within 4-6 months in 50-60 % of couples. A
strategy with diagnostic laparoscopy and LOD as first line of treatment of infertility in women with
PCOS will shorten the time to pregnancy for many women, reduce the need for medical ovulation
induction, and enable diagnosis of those women with anatomic infertility, who can only achieve
pregnancy by IVF treatment.
The first established surgical treatment for women with PCOS and infertility was ovarian wedge
resection1 . The dominant theory was that the pathological change in the ovaries was a result of a
basic pituitary hormone deficiency. This led to a lack of follicle ripening and ovulation, which led
to development of a multicystic disease with chronic thickening of the tunica. The more mature
follicles were believed destroyed intrinsically or by pressure obliteration and the younger, deeper
follicles were mechanically prevented from reaching the surface to mature and ovulate. The aim of
the operation was to assist the ripening of the remaining follicles by facilitating their approach to
The ovulation rate was high after surgery, and the pregnancy rate ranged from 25 to
86% in different studies2 . However, the incidence of postoperative periadnexal adhesion formation
was 25%; thus, it involved the risk of converting infertility due to an endocrine disorder to one with
a mechanical cause. The procedure is largely replaced by medical ovulation induction using
clomiphene citrate (CC) or follicle stimulating hormone/human chorionic gonadotropin (FSH/hCG)
as first line of treatment. The ovulation rate with CC is high, while the conception rate is 40 -50%
and the spontaneous abortion rate 30-40% in some studies3. Furthermore, 25% of the patients do not
respond to CC, and remain anovulatory despite increasing doses4 . Stimulation with CC or CC/FSH
is a stressful and expensive treatment requiring intense monitoring because of the associated risk of
a polyfollicular response followed by multiple pregnancies and the risk of ovarian hyper stimulation
A revived surgical approach to patients with PCOS and infertility is laparoscopic
ovarian drilling (LOD), introduced by Gjonnaess in 1984 5. The technique is used in different
surgical settings and most studies report the result of LOD in selected populations, as CC resistant
patients or concomitantly with CC and FSH treatment. The question is whether LOD should be first
line of treatment, second line of treatment after patients have proven resistant to CC, or third line of
treatment after failed ovulation induction with gonadotropins.
In this study we present the pregnancy rates after LOD in our department and evaluate
the optimal timing of LOD together with the other treatment regimens of infertile women with
The study group consisted of 57 women with PCOS referred with infertility to the Department of
Gynaecology at Holstebro Hospital, Denmark, during the period September 1, 1996 to September 1,
2002. Infertility was of minimum 12 months duration. PCOS was defined as the presence of
amenorrhoea, oligomenorrhoea or irregular menstruation with anovulation, and at least one of the
two following criteria: raised plasma-testosterone (> 1.8 nmol/l) or an LH/FSH ratio >3, if no
androgen hormone status was measured. Information about age, duration of infertility, former
pregnancies, former infertility treatment and menstrual cycle pattern was obtained as well as
information about previous infections, contraception and consumption of medicine and tobacco.
Oligomenorrhoea was defined as cycle duration between 35 days and 6 months, and amenorrhoea
was defined as the absence of menstruation for more than 6 months. The body mass index (BMI)
was calculated and plasma levels of LH, FSH, testosterone, dihydroepiandrosterone sulphate
(DHAS), sexual hormone binding globulin (SHBG), thyroid stimulating hormone (TSH) and
prolactin were measured in early follicular phase or at random in amenorrhoic women. Progesterone
was measured one week before expected menstruation or at random in amenorrhoic women. A
physical examination noting the presence of acne and amount and distribution of body hair was
performed, including a gynaecological examination and a vaginal ultrasound.
If the women were overweight defined as BMI > 29 kg/m2 they were offered dietician
advice for controlled weight loss. A semen sample from the male partner was analysed according to
the WHO criteria. The fertility was evaluated as followed based on the number of sperm cells with
normal motility: normal fertility > 10 millions/ ejaculate, reduced fertility = 2-10 millions/ ejaculate
and severe reduced/ non-existing fertility < 2 millions/ejaculate. Women with infertility and normal
or sufficient male fertility for insemination were included in this study. Women whose partners had
severely reduced fertility were excluded.
A hysteroscopy and a laparoscopy were routinely performed in all women within three
months after referral. Laparoscopy was performed via three ports of entry: a 10 mm laparoscope
inserted in the primary sub-umbilical trocar with two additional 5 mm trocars in the lower abdomen.
A grasping forceps was used to hold the ovarian ligament for manipulation of the ovary; the
diathermy needle was introduced via the other secondary port. Eight to ten holes were made in the
ovarian capsule with a unipolar coagulation current. Tubal patency was assessed by flushing the
tubes with methylene blue. Women with blocked tube or tubes were excluded from this study.
Women with non-obstructive tubo-peritoneal factors, e.g. slight peri-tubal or peri-ovarian adhesions
The management of infertility treatment after LOD was discussed with the couple and
in most cases expectant management was agreed upon i.e. three months pause before ovulation
induction was instituted. In women who postoperatively did not achieve regular menstrual cycles
and plasma progesterone of ≥ 30 mmol/l one week before the menstruation, stimulation with 100
mg of CC from the third to seventh cycle day was instituted and continued for a maximum of 3
cycles. If no pregnancy occurred, defined by positive urine hCG and the presence of an intrauterine
pregnancy verified by a vaginal ultrasound, intra-uterine insemination (IUI) was performed for a
maximum of 4 cycles during continued ovulation induction with CC and hCG. If the stimulation of
the ovaries was unsuccessful and ovulation did not take place, the hormonal treatment was changed
from CC to FSH starting with 75 IE daily on cycle day 3. If pregnancy did not occur using this
regimen, the women were referred to in-vitro fertilization (IVF).
Fischer´s Exact Test was applied to test independence between groups. Observation time was
calculated using the Kaplan-Meier method. The difference in observation time was analysed with
the Peto-Wilcoxon test. The difference between means were evaluated by Student´s t-test if data
followed Gaussian distribution, otherwise Mann-Whitney test was applied. The data in Table 2 was
tested with Chi-square - test for trend for contingency tables. The statistical software program,
SOLO (BMPD statistical packages, Berkeley CA, U.S.A.), was used for the calculations. Data are
given in median unless otherwise indicated. A two-sided value of p < 0.05 was chosen as the level
The crude pregnancy rate was 61% (35/57) and was achieved with LOD only in 32%
and followed by ovulation induction by gonadotropins in 29% of the women, respectively. The
crude pregnancy rate was 50% (6/12) among women with PCOS who had LOD as their only
treatment while 38% (17/45) became pregnant after combined treatment; 12 of 17 women who
became pregnant conceived spontaneously in an unstimulated cycle and had previously been treated
with CC/FSH (n=8) or CC/FSH + IUI (n=4) after LOD. No difference was observed in the clinical
preoperative data between the women who became pregnant and those who did not (Table 1).
Similar values were observed when the women were stratified according to preoperative treatment
with CC (data not shown).
The median time to pregnancy after LOD was 135 days (20 - 498) (Figure 1). The
time to pregnancy was similar for LOD only compared to postoperative ovulation induction (127,
60, and 191 days for women, who received no ovulation induction, CC/FSH only postoperatively,
and CC/FSH + IUI, respectively, Peto-Wilcoxon, Figure 2). When we stratified for treatment with
CC before LOD, no difference was found in time to pregnancy between previously CC treated and
untreated women (78, 122, and 185 days in CC resistant, CC sensitive and not preoperatively tested
women, respectively, Peto-Wilcoxon). Women who did not become pregnant had a median
observation time of 242 days (137 - 722) after LOD. The pregnancy rate depends on which time
after LOD the evaluation takes place (Figure 1).
Twenty-eight women (49%) had been treated with CC or CC + IUI prior to surgery.
Eleven of these women (39%) of the preoperatively treated women became pregnant after LOD
without stimulation treatment. Forty-three percent (12/28) were CC resistant before the operation
(Table 2) and 42 percent (5/12) of these became pregnant after LOD. No significant difference was
found in pregnancy rate in CC resistant women compared with CC sensitive women as well as those
women who had no preoperative treatment, respectively (p<0.11, Chi-square - test for trend, Table
Of the 35 pregnant women, one had a twin pregnancy, and six aborted spontaneously;
one of these fetuses had a confirmed Turner’s syndrome. In the women who got pregnant after LOD
+ CC, two cycles with CC (median) was sufficient. Those women who did not get pregnant on LOD
+ CC/FSH and subsequently had CC/FSH + IUI became pregnant after median one cycle. Twenty-
two women did not achieve pregnancy in our regimen. Six of these women are in current ovulation
induction and four women in expectant management. Six other women were referred to an IVF
clinic, and 2 of these subsequently became pregnant.
An often discussed issue in studies on LOD is which control group to use for comparison and that a
pregnancy rate depends on the time after LOD where evaluation takes place. Our data showed not
unexpectedly that pregnancy may occur as late as one year after LOD. Evaluation of pregnancy
rates involves observation over a certain time during which different treatment regimens may
influence the outcome in a differential manner. The vertical distance between the two curves in
Figure 1 shows those women who had not become pregnant at that particular time. While the
percent of women who do not get pregnant seems similar throughout the figure, this group,
however, does not consist of the same women at the start and at the end of the observation period.
Differences in observation time due to continuing inclusion of women for treatment and exclusion
due to stop of treatment of any reason add bias to the analysis of data if given as crude pregnancy
rates in a 2x2 table, for which reason the cumulated curve and the curve for the three different
postoperative treatment are presented. Furthermore, we found no differences in efficacy between
the expectant treatment after LOD and immediate ovulation induction. This gives sense to an
approach of a pause between the operation and medical treatment with gonadotropins. As indicated
in Figure 2 ovulation induction was instituted early in some women, which means that the effect of
treatment with LOD tends to be underestimated. However, long-term studies indicate that the effect
of LOD may level off and in only one third of the women regular menstrual cycles last for more
than 3 years6. Adding to this phenomenon, the underlying endocrinopathy in PCOS is less
pronounced as women get older and therefore cycles are relatively more regular6-8. However, this
may not increase conception rate but overestimate the effect of LOD on menstrual patterns and
No difference in clinical characteristics was found between the women who,
irrespective of treatment, got pregnant and the women who did not. Therefore, we cannot identify
which women who would benefit from one treatment or the other. One large study of LOD found
duration of fertility and LH levels preoperatively correlated with pregnancy rate, while drilling with
diathermy performed better than with laser 9. However, no information on postoperatively ovulation
induction was given.
PCOS is not a well-defined disease but a condition characterized by a variety of
symptoms. The lack of consensus on the definition of PCOS makes comparison between studies
difficult. In a Cochrane review on the effectiveness of LOD in the treatment of PCOS, it was
concluded that the number of randomised controlled studies were too small and the quality of the
studies too poor to draw a conclusion10 . Nevertheless, results from cohort studies indicate that there
may be some effect of LOD6,9. Most studies include various postoperative ovulation induction
regimes introduced at various times to women who previous failed to ovulate or conceive with
stimulation. Thus, the potential effect of LOD alone in PCOS women was difficult to discern from
other forms of treatment11,12. In our study, the laparoscopy was used not for the purpose of LOD
alone but as a diagnostic tool to exclude other causes of infertility. This saved those women
repeated ovulation induction treatment which had no prospect of succeeding with this treatment.
Secondly, other surgical procedures with the potential to increase fertility could be performed, like
removing sactosalpinx or releasing adhesions.
Ovarian drilling is believed to be a relatively simple procedure adding only minimal
morbidity to the diagnostic laparoscopy, but there are potential disadvantages and complications to
be considered9 . The incidence of postoperative intra-peritoneal adhesion formation has been found
to be 19-26%5,11-13. Excessive drilling close to the ovarian hilus and potentially damaging the
ovarian blood supply should be avoided, since it may lead to premature ovarian failure5.
In this study, we have included women with proven CC resistance, women who
ovulated after CC treatment, and women who had not been treated with CC at all preoperatively.
We could not find any conclusive difference in pregnancy rate between these groups to favour LOD
for any particular group compared to the other; the reason may be that the groups are too small.
Other studies have shown effect of LOD in previous CC resistant women with PCOS mainly
comparing with medical induction alone6,9,12,14. However, if ovulation induction is instituted before
LOD, one would unknowingly treat women with tubal factor and postpone efficient treatment for
these women. The concern of potential short and long-term effects of ovulation induction treatment
has been raised15. There is currently no consensus for how long one may treat infertile women with
gonadotropins but it makes sense to weigh the continuing medical treatment against the benefits of
a preoperative treatment of PCOS combined with the diagnosis or exclusion of an anatomic reason
We have evaluated the number of women achieving pregnancy with respect to whether
medical ovulation induction was sought afterwards. This is in line with studies evaluating treatment
of infertility evaluating other factors important for the couple, like economy, mental stress, and
potential future health consequences for the woman. In a British study, the relative cost-
effectiveness of different treatments for infertility was assessed in relation to probability of clinical
pregnancy, complications, cancellation of cycles and couple’s withdrawing before the end of the
course of treatment. It concluded that LOD and medical treatment was cost-effective for infertile
women with PCOS 16. Still, it is not clear whether subgroups of the very heterogeneous population
of PCOS women would benefit more from other forms of non-invasive treatment, e.g. glitazones or
metformin. The latter is reported to increase ovulatory and pregnancy rate when combined with CC
in women with PCOS17,18.
In conclusion this study showed that LOD alone resolves the infertility within 4-6
months in 50-60 % of couples. A strategy with diagnostic laparoscopy and LOD as first line of
treatment of infertility in women with PCOS will shorten the time to pregnancy for many women,
reduce the need for medical ovulation induction and enable diagnosis of those women with
anatomic infertility, who can only achieve pregnancy by IVF treatment. The row of sequence of the
outlined strategy is that other causes of infertility are ascertained and LOD performed before
stimulation is instituted.
Table 1. Clinical data of 57 women with PCOS treated with LOD.
Not pregnant after Pregnant after LOD
LOD In unstimulated cycles In stimulated cycles
Number 22 18 17
Age (years) 28(21-38) 27(21-31) 27(22-33)
BMI (kg/m ) 29(20-43) 26(18-34) 26(18-34)
Duration of infertility
29(12-84) 20(12-42) 21(12-78)
LH (IU/l) 13(4-31) 16(5-48) 14(5-24)
FSH (IU/l) 5(1-8) 5(2-8) 6(4-8)
LH/FSH ratio 2.8(0.5-6.2) 3.3(1-7.5) 2.3(1-4.5)
Testosterone (nmol/l) 3(1.6-8.4) 2.9(1.9-4.1) 3.9(1.2-14)
SHBG (nmol/l) 42(8-121) 42(14-114) 49(17-120)
DHAS (nmol/l) 5801(2883-11220) 6501(3700-12477) 5728(1900-9592)
Primary infertility 17 12 14
Secondary infertility 7 5 2
Amenorrhoea 9 2 6
Oligomenorrhoea 8 11 7
Elevated androgens 21 17 15
Androgens not measured 1 1 2
CC resistance§ 7/14 1/8 4/6
Sperm count (106/ml) 64(14-159) 79(12-210) 81(8-165)
Data are given as numbers and mean (range). Normal ranges: Testosterone 0.55-1.8 nmol/l, SHBG
41-169 nmol/l, DHAS 1200-9500 nmol/l. §: number of women with CC resistance / number of
women tested before LOD.
Table 2. Incidence of pregnancy in women after LOD in relation to preoperative clomiphene
treatment. Data given as numbers (percent).
LOD + Total no. of
LOD alone CC/FSH +
1 2 2 5 (42)
CC resistant (n = 12)
8 1 1 13 (81)
CC sensitive (n = 16)
9 4 7 20(69)
18 7 10 35(61)
1. Stein I.F., Cohen M.R. Surgical treatment of bilateral polycystic ovaries-amenorrhea and sterility.
Am J Obstet Gynecol 1939;38:465-73
2. Tulandi T, Took S.A. Surgical management of polycystic ovarian syndrome. Baillière Clin
Obstet Gynecol 1998;12:41-553
3. Felemban A, Lin Tan S, Tulandi T. Laparoscopic treatment of polycystic ovaries with insulated
needle cautery: a reappraisal. Fertil Steril 2000;73:266-9
4. Tulandi T, Watkin K, Lin Tan S. Reproductive performance and three-dimensional ultrasound
volume determination of polycystic ovaries following laparoscopic ovarian drilling. Int J Fertil
5. Gjonnaess H. Polycystic ovarian syndrome treated by ovarian electrocautery through the
laparoscope. Fertil Steril 1984;41(1):20-5
6. Amer SAK, Goplan V, Li TC, Ledger WL, Cooke ID. Long term follow-up of patients with
polycystic ovarian syndrome after laparoscopic ovarian drilling: clinical outcome. Hum Reprod
7. Dahlgren E, Johansson S, Lindstedt G et al. Women with polycystic ovary syndrome wedge
resected in 1956 to 1965: a long-term follow-up focusing on natural history and circulating
hormones. Fertil Steril 1992:57:505-13.
8. Elting MW, Korsen TJM, Rekers-Mombarg LTM et al. Women with polycystic ovary syndrome
gain regular menstrual cycles when ageing. Hum Reprod 2000;15:24-8
9. Li TC, Saravelos H, Chow MS et al. Factors affecting the outcome of laparoscopic drilling for
polycystic ovarian syndrome in women with anovulatory infertility. Br J Obstet Gynaecol
10. Farquhar C, Vandekerckhove P, Arnot M et al. Laparoscopic ”drilling” by diathermy or laser
for ovulation induction in anovulatory polycystic ovary syndrome (Cochrane Review). In: The
Cochrane Library, Issue 1, 2001. Oxford: Update Software.
11. Liguori G Tolino A, Moccia G et al. Laparoscopic ovarian treatment in infertile patients with
polycystic ovarian syndrome (PCOS): endocrine changes and clinical outcome. Gynecol
12. Naether OGJ, Baukloh V, Fischer R et al. Long term follow-up in 206 infertility patients with
polycystic ovarian syndrome after laparoscopic electrocautery of the ovarian surface. Hum
13. Naether OGJ, Fischer R, Weise HC et al. Laparoscopic electrocoagulation of the ovarian
surface in infertile patients with polycystic ovarian disease. Fertil Steril 1993;1:88-94.
14. Farquhar CM, Williamson K, Gudex G et al. A randomized controlled trial of laparoscopic
ovarian diathermy versus gonadotropin therapy for women with clomiphene citrate-resistant
polycystic ovary syndrome. Fertil Steril 2002;78:404-11.
15. Rossing MA, Daling JR, Weiss NS et al. Ovarian tumours in a cohort of infertile women. N
Engl J Med 1994;331:771-6.
16. Philips Z, Barraza-Llorenz M, Posnett J. Evaluation of the relative cost-effectiveness of
treatments for infertility in the UK. Hum Reprod 2000;15(1):95-106.
17. Vandermolen DT Ratts VS, Evans WS et al. Metformin increases the ovulatory rate and
pregnancy rate from clomiphene citrate in patients with polycystic ovary syndrome who are
resistant to clomiphene citrate alone. Fertil Steril 2001;75:310-5
18. Kocak M, Caliskan E, Simsir C et al. Metformin therapy improves ovulatory rates, cervical
scores, and pregnancy rates in clomiphene citrate-resistant women with polycystic ovary
syndrome. Fertil Steril 2002;77:101-6
Cumulated observation time after LOD in 57 women with PCOS. Observation time halted when the
woman either became pregnant, further treatment ceased or data entry for study stopped.
Broken line: All 57 women. Full line: The 35 women who became pregnant.
Cumulated pregnancy rate in 35 women with PCOS by treatment regimen.
Full line: LOD only (n = 18). Full line with filled circles: Pregnancy in LOD + CC/FSH (n = 7).
Broken line with filled triangles: Pregnancy in LOD +CC/FSH+ IUI cycle (n = 10).
Pregnancy rate in LOD + CC/FSH + IUI cycle versus LOD + CC/FSH cycle, p< 0.05, Peto-
Pregnancy rate (%)
0 50 100 150 200 250 300 350 400 450 500
Days after drilling
Pregnancy rate (%)
0 50 100 150 200 250 300 350 400 450 500
Days after drilling