Human Reproduction Vol.17, No.4 pp. 1046–1051, 2002

Laparoscopic ovarian diathermy in women with polycystic ovarian syndrome: a retrospective study on the influence of the amount of energy used on the outcome S.A.K.Amer1, T.C.Li and I.D.Cooke The University of Sheffield, Jessop Wing, Sheffield Teaching Hospitals, Tree Root Walk, Sheffield S10 2SF, UK 1To

whom correspondence should be addressed. E-mail: [email protected]

BACKGROUND: Currently, there is an uncertainty about the optimum number of punctures to be applied at laparoscopic ovarian diathermy (LOD). This retrospective study was undertaken to investigate the dose–response relationship of LOD. METHODS: The hospital records of 161 women with polycystic ovarian syndrome who underwent LOD were reviewed and the clinical data before and after LOD were documented. Subjects were divided into six groups according to the number of punctures made in their ovaries as follows: group 1, two punctures per ovary; group 2, three punctures; group 3, four punctures; group 4, five punctures; group 5, six punctures and group 6, seven to 10 punctures. Contingency table analysis and analysis of variance were used to compare the outcomes of the different groups. RESULTS: The rates of ovulation, conception and restoration of menstrual regularity after LOD were significantly lower in group 1 compared with other groups. There were no significant differences in the success rates between the other groups. CONCLUSION: Two punctures per ovary are associated with poor results. Three punctures per ovary seem to represent the plateau dose. The application of seven or more punctures per ovary may result in excessive destruction to the ovary without any improvement of the results and should therefore be discouraged. Key words: laparoscopic ovarian diathermy/polycystic ovarian syndrome/polycystic ovaries

Introduction Although laparoscopic ovarian diathermy (LOD) is widely practised by many gynaecologists as an effective secondline treatment for clomiphene citrate-resistant anovulatory infertility associated with polycystic ovarian syndrome (PCOS), there is a lack of consensus on how much energy should be used. It depends on the number of punctures made, power setting and duration of each puncture. Currently, the choice of the number of punctures to be applied at LOD is empirical. Historically, the amount of ovarian tissue removed during bilateral ovarian wedge resection varied between onethird to three-quarters depending on the size of the ovary (Stein, 1964; Judd et al., 1976; Loppo¨hn and Bogchelman, 1989). Some gynaecologists apply the same principle to LOD by empirically making different numbers of punctures in each ovary depending on its size (Naether et al., 1993; Merchant, 1996; Li et al., 1998; Tulandi, 1999; Felemban et al., 2000). Between three and 25 punctures have been reported with power settings between 30 and 400 watts (Gjonnaess, 1984; Dabirashrafi, 1989; Weise et al., 1991; Naether et al., 1993; Felemban et al., 2000). As a general principle, increasing the amount of thermal energy delivered to the ovarian stroma may 1046

increase the efficacy of the procedure but at the expense of increasing the risk of ovarian atrophy. Gjonnaess reported that ovulation occurred more frequently if five or more holes were applied compared with three holes per ovary (Gjonnaess, 1984). However, Armar et al. found that four diathermy holes per ovary were sufficient to achieve good results and that no improvement was achieved when applying more holes (Armar et al., 1990). Weise et al. noted a positive correlation between the number of holes and the decrease in serum testosterone concentrations following LOD (Weise et al., 1991). However, there are a number of difficulties in comparing the experience of various authors due to variation in the techniques used in LOD, including: (i) using different instruments (needles, scissors, biopsy forceps, etc.) to deliver the energy to the ovary; (ii) applying different amount of energy to the ovary (measured in joules, equivalent to power in watts multiplied by the duration of electricity applied in s per puncture); and (iii) distribution of the thermal energy, either localized to a few holes or more widely spread over many holes with varying depths of penetration. In this retrospective study we wish to investigate the dose– response relationship of LOD by analysing the influence of © European Society of Human Reproduction and Embryology

Energy usage and outcome of LOD

the amount of thermal energy delivered to each ovary on the clinical outcome. Materials and methods Subjects Between 1991 and 2000, a total of 221 women with anovulatory infertility associated with PCOS underwent laparoscopic ovarian drilling in our unit. The hospital records of these 221 subjects were reviewed. Thirty-eight women had ovarian drilling using laser and were excluded. The remaining 183 women were treated with ovarian diathermy. In 22 of these subjects, the number of punctures made in the ovaries during LOD was not clearly recorded in their hospital records and they were therefore excluded. The remaining 161 patients provided the data for our study. All the women had anovulatory infertility of ⬎1 year duration, and had been unsuccessfully treated with clomiphene citrate of up to 150 mg/day for 5 days in the early follicular phase of the menstrual cycle prior to LOD. In 128 women, clomiphene citrate failed to induce ovulation (clomiphene citrate resistance). The remaining 33 patients ovulated but failed to conceive after clomiphene citrate treatment for 6–9 months. In addition, 11 women had received HMG therapy for ovulation induction and failed to conceive. A diagnosis of PCOS was based on the following criteria: (i) early follicular phase (defined as days 2–5 of the menstrual cycle) serum LH:FSH ratio was ⬎2 and/or raised serum androgen levels [testosterone 艌2.5 nmol/l, androstenedione 艌10 nmol/l or free androgen index (FAI) ⬎4]. FAI has been found to be a useful test for detecting patients with PCOS (Eden et al., 1989). It is calculated using the formula: testosterone⫻100/sex hormone-binding globulin (SHBG) (Carter et al., 1983). In women who were oligo-/amenorrhoeic, a random blood sample was acceptable; or (ii) there was ultrasonographic evidence of ovarian stromal hypertrophy and multiple (艌10), small (6–8 mm) follicles arranged in the periphery of the ovary (Adams et al., 1985). Laparoscopic ovarian diathermy The techniques of laparoscopic ovarian drilling used in our centre have been previously published (Li et al., 1998). Over 10 gynaecologists of varying degrees of experience (senior house officers under supervision, registrars and consultants) carried out the treatment over the10 year period. In most cases a three-puncture laparoscopy was performed. A 10 mm laparoscope was inserted via the main subumbilical entry and a pair of grasping forceps was introduced through one of the two lower abdominal 5 mm punctures to grasp the utero–ovarian ligament and to lift the ovary away from the bowel. The third entry was used to introduce the diathermy needle. A specially designed probe (Rockett of London, Watford, UK) was used to penetrate the ovarian capsule to a standard depth (8 mm) at a number of points with the aid of a short burst of diathermy. The probe has a distal stainless steel needle measuring 8 mm in length and 2 mm in diameter and projecting from an insulated solid cone of 6 mm maximum diameter. The electrosurgical unit used was the Force 2 Valleylab® electrosurgical generator (Valleylab Inc., Boulder, CO, USA). A monopolar coagulating current at 30 watt power setting was used and the duration of each penetration was about 5 s. A total of two to 10 punctures were made in each ovary depending on its size. Each ovary was cooled by irrigation using Hartmann’s solution before releasing the ligament. At the end of the procedure, ~200 ml of Hartmann’s solution were left in the pelvis. Post-operative monitoring Following ovarian diathermy, women were asked to keep a record of their menstrual cycle. If the patient started a menstrual period within

Table I. The characteristics of 161 women who underwent laparoscopic ovarian diathermy for anovulatory infertility due to polycystic ovarian syndrome. Values are given as mean ⫾ SD (range) and number of observations as n (%). The results shown are those obtained before the operation Characteristic Age (years) Body mass index (kg/m2) Duration of infertility (years) Serum LH (IU/l) Serum FSH (IU/l) Serum LH:FSH ratio Serum testosterone (nmol/l) (n ⫽ 88) Serum androstenedione (IU/l) (n ⫽ 75) Free androgen index (n ⫽ 72) Ovarian volume (ml3) (n ⫽ 43) Menstrual cycle pattern Regular Oligomenorrhoea Amenorrhoea Hirsutism/acne Yes No Infertility Primary Secondary

Results 29.2 ⫾ 4.0 (20–39) 27.1 ⫾ 5.0 (18–41) 3.3 ⫾ 2.2 (1–12) 14.3 ⫾ 7.1 (1.8–37) 5.3 ⫾ 1.6 (2.3–7.2) 2.8 ⫾ 1.3 (0.5–6.2) 2.5 ⫾ 1.3 (0.7–5.9) 9.4 ⫾ 4.1 (1.7–19.2) 8.2 ⫾ 6.7 (1.0–37) 11.1 ⫾ 3.7 (4.3–19.5) 12 (7%) 112 (70%) 37 (23%) 53 (33%) 108 (68%) 104 (65%) 57 (35%)

6 weeks of the surgery, a blood sample was taken on day 2 of that cycle for measurement of serum concentrations of LH, FSH, testosterone, androstenedione and SHBG. Another blood sample was taken on day 21 of the same cycle for measurement of serum concentration of progesterone. Ovulation was diagnosed when the progesterone level was 艌30 nmol/l. Two more mid-luteal phase blood samples were taken in the subsequent cycles to measure serum progesterone levels. If spontaneous menstruation did not occur during the 6 weeks following surgery, a random blood sample was taken to measure all the above hormones. Analysis of the data The clinical and biochemical data before and after LOD were documented from the hospital records. The age of the patients and other demographic details including body mass index (BMI), primary or secondary infertility, and the duration of infertility were also documented. Patients were divided into six groups according to the number of punctures made in their ovaries during LOD as follows: group 1 (n ⫽ 12) treated with two punctures per ovary, group 2 (n ⫽ 15) three punctures, group 3 (n ⫽ 27) four punctures, group 4 (n ⫽ 43) five punctures, group 5 (n ⫽ 25) six punctures and group 6 (n ⫽ 39) seven to 10 punctures. The clinical data were entered into the Statistical Package for Social Science for PC version 10.0.5. Appropriate statistical tests including contingency table analysis and analysis of variance were used.

Results The demographic, clinical and endocrinological characteristics of the 161 women who underwent LOD are shown in Table I. In Tables IIA and IIB, the characteristics of each group are shown separately. Analysis of variance showed no difference among the six groups except that group 3 (four punctures) had significantly lower mean pre-operative LH (11.0 nmol/l; 1047

S.A.K.Amer, T.C.Li and I.D.Cooke

Table II. The characteristics of 161 women with PCOS who underwent laparoscopic ovarian diathermy: comparison between women treated with different doses of thermal energy (numbers of punctures). Values are given as mean (SEM). A. Demographical and clinical characteristics

Punctures (energy in J) per ovary Age Body mass index (kg/m2) Duration of infertility Menstrual cycle Regular Oligomenorrhoea Amenorrhoea

Group 1 (n ⫽ 15)

Group 2 (n ⫽ 12)

Group 3 (n ⫽ 27)

Group 4 (n ⫽ 43)

Group 5 (n ⫽ 25)

Group 6 (n ⫽ 39)

2 (300) 28.7 (1.0) 25.8 (1.1) 3.0 (0.44)

3 (450) 29.7 (1.5) 29.3 (1.7) 2.6 (0.3)

4 (600) 29.3 (0.9) 26.1 (1.0) 3.1 (0.4)

5 (750) 28.9 (0.6) 27.8 (0.8) 3.2 (0.3)

6 (900) 29.5 (0.8) 28.2 (1.0) 2.9 (0.3)

艌7 (艌1050) 29.0 (0.6) 26.2 (0.7) 3.9 (0.5)

1 (7%) 9 (60%) 5 (33%)

1 (8%) 9 (75%) 2 (16%)

3 (11%) 18 (67%) 6 (22%)

3 (7%) 29 (67%) 11 (26%)

2 (8%) 20 (80%) 3 (12%)

2 (5%) 27 (69%) 10 (26%)

Group 1 (n ⫽ 15)

Group 2 (n ⫽ 12)

Group 3 (n ⫽ 27)

Group 4 (n ⫽ 43)

Group 5 (n ⫽ 25)

Group 6 (n ⫽ 39)

2 (300) 16.7 (2.1) (n ⫽ 15) 5.7 (0.4) (n ⫽ 15) 2.3 (0.4) (n ⫽ 12) 9.5 (1.4) (n ⫽ 12) 53.9 (5.7) (n ⫽ 12) 5.0 (0.9)a (n ⫽ 12) 12.0 (1.4) (n ⫽ 7)

3 (450) 14.9 (2.6) (n ⫽ 12) 5.8 (0.3) (n ⫽ 12) 2.0 (0.4) (n ⫽ 7) 7.1 (1.4) (n ⫽ 6) 41.5 (8.0) (n ⫽ 6) 7.0 (3.0) (n ⫽ 6) 10.1 (2.9) (n ⫽ 8)

4 (600) 11.0 (1.0)a (n ⫽ 27) 4.2 (0.3)b (n ⫽ 27) 2.5 (0.3) (n ⫽ 17) 9.7 (0.9) (n ⫽ 15) 43.5 (4.8) (n ⫽ 17) 7.2 (1.3) (n ⫽ 17) 11.6 (1.8) (n ⫽ 9)

5 (750) 13.8 (0.9) (n ⫽ 43) 5.6 (0.2) (n ⫽ 43) 2.3 (0.2) (n ⫽ 25) 9.7 (0.8) (n ⫽ 19) 39.8 (3.7) (n ⫽ 21) 7.5 (0.9) (n ⫽ 21) 11.7 (0.7) (n ⫽ 10)

6 (900) 12.7 (1.4) (n ⫽ 24) 5.5 (0.5) (n ⫽ 24) 2.6 (0.4) (n ⫽ 10) 9.0 (1.2) (n ⫽ 11) 37.4 (7.2) (n ⫽ 8) 11.7 (3.4) (n ⫽ 8) 9.6 (1.4) (n ⫽ 6)

艌7 (艌1050) 16.8 (1.3) (n ⫽ 36) 5.4 (0.3) (n ⫽ 36) 3.2 (0.4) (n ⫽ 17) 10.0 (0.9) (n ⫽ 12) 36.6 (4.8) (n ⫽ 11) 12.9 (3.3) (n ⫽ 11) 10.4 (5.9) (n ⫽ 2)

B. Endocrinological and ultrasonographical characteristics

Punctures (energy in J) per ovary LH (nmol/l) FSH (nmol/l) Testosterone (nmol/l) Androstenedione (nmol/l) SHBG (nmol/l) FAI Ovarian volume

⬍ 0.05; bP ⬍ 0.01 versus all other groups. J ⫽ joules; FAI ⫽ free androgen index; SHBG ⫽ sex hormone binding globulin.

aP

P ⬍ 0.05) and FSH (4.2 nmol/l; P ⬍ 0.01) levels compared with the other groups (LH, 12.7–16.8 nmol/l; FSH, 5.4–5.8 nmol/l). The FAI for group 1 (5.0) was significantly lower (P ⬍ 0.05) than that of all other groups. Ovulation rates Figure 1a shows the rate of spontaneous ovulation achieved using different doses of thermal energy. The rates of ovulation in groups 1–6 were 27, 58, 62, 60, 59 and 54% respectively. Contingency table analysis showed that the incidence of ovulation in group 1 was significantly lower (P ⬍ 0.05) than the other groups. Conception Figure 1b shows the conception rates achieved with different doses of thermal energy during the first year after surgery. Contingency table analysis showed that the pregnancy rate in group 1 (13%) was significantly lower (P ⬍ 0.05) than the other groups (45–60%). Menstrual pattern Figure 1c shows the proportion of women with regular menstrual cycles after ovarian diathermy. Contingency table analysis showed that the proportion of women with regular 1048

menstrual cycles after LOD in group 1 (27%) was significantly lower (P ⬍ 0.05) than the other groups (61–78%). Endocrine changes The magnitude of changes in the gonadotrophin levels after ovarian diathermy, using different amounts of thermal energy, is shown in Figure 2. Analysis of variance showed no difference between groups 1–5 in the percentage change of LH levels. LH/FSH ratio showed a 10% increase in group 1 compared with a reduction of 22–43% in the rest of the groups. The reduction in LH levels and LH/FSH ratio after LOD was highest (40 and 43% respectively) in group 6 (艌7 punctures per ovary). The changes of the serum levels of FSH after LOD were variable and insignificant in groups 1–5 (–10 to ⫹9%), but there was a significant increase of 29% in group 6. Discussion In this retrospective study, we have investigated the influence of the amount of thermal energy applied at LOD on the clinical and biochemical outcome in 161 women with PCOS. To the best of our knowledge, this is the largest series to be reported on the dose–response relationship of LOD. Seven percent of women in this study had apparently

Energy usage and outcome of LOD

Figure 1. The rates of spontaneous ovulation, conception and conversion of oligo-/amenorrhoea to regular cycles in women with polycystic ovarian syndrome after laparoscopic ovarian diathermy (LOD) using different doses of thermal energy (number of punctures). *P ⬍ 0.05.

Figure 2. The magnitude of decrease/increase of serum concentrations of gonadotrophins in women with polycystic ovarian syndrome after laparoscopic ovarian diathermy (LOD) using different amounts of thermal energy (number of punctures). Data show the percentage decrease/increase of the hormones and also the LH/FSH ratio. Analysis of variance showed no difference between groups in the percentage change of FSH and LH levels, but the change in LH /FSH ratio in group 1 (⫹10%) was significantly different (P ⬍ 0.05) from the rest of the groups (–22 to –43%).

regular menstrual cycles prior to LOD. Although chronic anovulation in women with PCOS is usually associated with menstrual irregularities (Franks, 1995), several authors have reported that a proportion of these women do have apparently ‘regular’ menstrual cycles. It was reported that 21% of anovulatory PCOS women have regular menstrual cycles (Carmina and Lobo, 1999). Similarly, a 24% incidence of regular menstrual cycles among anovulatory PCOS patients undergoing ovarian diathermy was reported (Naether et al., 1994). In a classic review of 187 reports describing 1079 cases of PCOS, a 16% incidence of regular menses was reported (Goldzieher and Axelrod, 1963). This observation was confirmed in a recent study of 1741 women with PCOS in which 30% of patients had regular menses (Balen et al., 1995). Furthermore, many anovulatory PCOS patients ovulate occasionally and some may resume regular menstrual cycles for variable periods of time. This explains why some anovulatory PCOS patients conceive spontaneously while being investigated for infertility or waiting for treatment.

Calculation of thermal energy Gjonnaess correlated the ovulation rates after LOD to different numbers of points cauterized (Gjonnaess, 1984). He reported that the best results [ovulation rate ⫽ 96.7% (n ⫽ 30)] were obtained when the number of points was more than five per ovary (⬎10 points per patient). However, this referred to the use of biopsy or sterilization forceps applied against the ovarian surface and activating the electricity at 200–300 watts for 3 ⫾ 1 s. Hence the amount of thermal energy delivered to each ovary, on average, was 250 watts⫻3 s⫻⬎5 ⫽ ⬎3750 joules. Armar et al. found that four diathermy holes per ovary were sufficient to achieve good results and that no improvement was achieved when applying more holes (Armar et al., 1990). This referred to the use of a specially designed needle to penetrate the ovary with activation of the electricity (40 watts) for 4 s at each point. The amount of thermal energy delivered to each ovary was 40 watts⫻4 s⫻4 ⫽ 640 joules, which is significantly lower than that used by Gjonnaess (Gjonnaess, 1984). In our study, the amounts of thermal energy used in each group are as follows. Group 1: 30⫻5⫻2 ⫽ 300 joules; 1049

S.A.K.Amer, T.C.Li and I.D.Cooke

group 2: 30⫻5⫻3 ⫽ 450 joules; group 3: 30⫻5⫻4 ⫽ 600 joules; group 4: 30⫻5⫻5 ⫽ 750 joules; group 5: 30⫻5⫻6 ⫽ 900 joules; and group 6: 30⫻5⫻艌7 ⫽ 艌1050 joules. It is important that the comparison between different studies should take into consideration the total amount of thermal energy delivered to each ovary, not just the number of holes made in the ovary. Threshold dose In our study, the application of two holes per ovary, equivalent to the delivery of 300 joules, was found to produce significantly poorer results than the other groups, measured in terms of restoration of menstrual regularity, ovulation rate and conception rate. While it is possible that two punctures (300 joules) per ovary represent the threshold dose (i.e. the lowest dose at which a response could be seen), it is of interest to note that the LH/FSH ratio did not decrease in group 1 (treated with two punctures per ovary), suggesting that the responders in this group could represent treatment-independent events. The threshold dose may therefore be higher than two punctures per ovary. Plateau dose The plateau dose refers to the lowest dose at which all subjects who will respond are observed to respond. In our study, there did not appear to be significant differences in the outcomes (menstrual pattern, ovulation and conception) of groups 2–6. It seems that three punctures (450 joules) per ovary produce results as good as higher numbers of punctures. However, this is a retrospective study, and caution should be exercised in the interpretation of results. Firstly, it is possible that there was selection bias by the surgeons in the choice of the number of punctures. One example is that surgeons might have applied more punctures to larger ovaries. Secondly, our own analysis suggested that LH and FSH levels were lower in group 3 (four punctures per ovary). There is no obvious explanation for such a finding, apart from some form of selection bias or purely a chance occurrence. The lower LH levels may also indicate that women in this group had a milder form of the syndrome and this might have an impact on their response to the treatment. Thirdly, in a retrospective study such as this one, while the number of punctures made and the power setting could be accurately quantified, the duration of application of thermal energy (5 s) is only approximate. Consequently, our study only provides a rough guide to the dose of thermal energy used. A prospective dose-finding study is required to estimate more accurately the optimal amount of energy required for LOD. Excessive thermal energy Ovarian atrophy and failure is a rare complication of LOD. Dabirashrafi reported a case of severe ovarian atrophy following LOD in which eight punctures were created at 400 watts for 5 s per each puncture, equivalent to 16 000 joules (Dabirashrafi, 1989). It is therefore possible that application of excessive amounts of thermal energy to the ovary during LOD will produce irreversible damage to the ovary, leading to ovarian failure. 1050

FSH is considered to be a reasonable marker of ovarian reserve and function. In our study, it is of interest that the application of seven or more punctures (艌1050 joules) per ovary resulted in a 29% increase in the FSH levels after LOD, compared with –10 to 9% change in the other groups. It seems possible that the application of seven or more punctures per ovary represents an excessive amount of thermal energy used for LOD and should therefore be discouraged. Depth of penetration of energy Gjonnaess used a significantly greater amount of thermal energy (⬎3750 joules per ovary) than that used in our study (Gjonnaess, 1984). However, Gjonnaess, using a pair of biopsy or sterilization forceps, applied the thermal energy to the ovarian surface and the depth of penetration was therefore between 2–4 mm, i.e. superficial. In contrast, the depth of penetration in our study using the specially designed ovarian diathermy needle (Rocket of London) was up to 8 mm. A similar instrument was used by others, who found that the amount of thermal energy used to produce a good result was ~640 joules per ovary, which is similar to that found in our study (Armar et al., 1990). Similarly, Felemban et al. delivered the energy to a depth of 8 mm and achieved good success rates (73% ovulation and 54% conception rates) with amounts of thermal energy (40 watts⫻2 s⫻10–15 ⫽ 800–1200 joules) significantly lower than those applied by Gjonnaess (Gjonnaess, 1984; Felemban et al., 2000). It is therefore possible to conclude that with deeper penetration during LOD, the amount of thermal energy can be reduced without compromising the outcome. Furthermore, achieving good results with deep penetration using low energy supports the hypothesis that LOD works by destroying androgen-producing ovarian stroma. Endocrine outcome It is of interest to note that the magnitude of decrease in LH levels and LH/FSH ratio after LOD seems to be dosedependent, with the highest reduction being achieved in group 6 (seven or more punctures per ovary). This greater reduction of LH levels could be explained by the greater reduction of serum androgen concentrations as a result of greater destruction of the androgen-producing ovarian stroma with higher doses of thermal energy. This in turn results in reduction of the peripheral aromatization of androgens to estrone. The resulting fall in estrone may be responsible for decreased positive feedback on LH and decreased negative feedback on FSH at the level of the pituitary. However, the greater reduction of the serum LH levels achieved with higher doses of thermal energy was not associated with an increase of the ovulation and pregnancy rates. This indicates that mechanisms of actions other than that described above may be responsible for the effects of LOD. For instance, it has recently been suggested that the ovary produces a number of growth factors, such as insulin-like growth factor-I in response to tissue injury, which sensitize the ovary to circulating FSH resulting in stimulation of follicular growth. It has also been hypothesized that minimal ovarian injury leads to the production of non-steroidal factors which affect the ovarian–pituitary feedback resulting in an attenuated response of LH secretion to stimulation with GnRH,

Energy usage and outcome of LOD

which leads to a decrease in serum LH concentrations (Rossmanith et al., 1991). Although groups 2, 3, 4 and 6 showed post-operative increase of the mean serum levels of FSH, group 5 unexpectedly showed a reduction of FSH levels. This resulted in a lower than expected magnitude of post-operative reduction of LH/FSH ratio in this group. As mentioned above, the choice of the number of punctures to be made during ovarian diathermy was based on the ovarian size. Therefore, it is possible that women in groups 5 and 6 had larger ovaries, which may indicate a severer form of the syndrome. It is possible that the amount of thermal energy used in group 5 (900 joules per ovary) was not enough to produce an increase of FSH levels due to the relatively larger size of the ovaries in this group. In group 6, the amount of energy used (1050– 1500 joules) was significantly greater than that of group 5 and therefore the endocrine changes were greater. The serum androgen levels and ovarian volumes were not routinely measured after LOD and therefore the available data were limited and insufficient for analysis. In conclusion, in this retrospective study we analysed the influence of the number of punctures made at LOD on the clinical outcome in 161 women with PCOS. Three punctures (450 joules) per ovary seemed to be the plateau dose for LOD. Deeper penetration with the specially designed diathermy needle allows reduction of the amount of thermal energy delivered to the ovary without affecting the success rates. Making seven or more punctures per ovary appears to deliver an excessive amount of thermal energy to the ovary. This offers no advantage over the lower doses of energy in terms of success rates, and may potentially cause excessive ovarian damage. There is a need for a prospective dose-finding study in order to estimate the optimal amount of thermal energy required for LOD.

References Adams, J., Franks, S., Polson, D.W., Mason, H.D., Abdulwahid, N., Tucker, M., Morris, D.V., Price, J. and Jacobs, H.S. (1985) Multifollicular ovaries: clinical and endocrine features and response to pulsatile gonadotrophin releasing hormone. Lancet, ii, 1375–1378. Armar, N.A., McGarrigle, H.H.G., Honour, J., Holownia, P., Jacobs, H.S. and Lachelin, G.C.L. (1990) Laparoscopic ovarian diathermy in the management of anovulatory infertility in women with polycystic ovaries: endocrine changes and clinical outcome. Fertil. Steril., 53, 45–49.

Balen, A.H., Conway, G.S., Kaltsas, G., Techatrasak, K., Manning, P.J., West, C. and Jacobs, H.S. (1995) Polycystic ovary syndrome: the spectrum of the disorder in 1741 patients. Hum. Reprod., 10, 2107–2111. Carmina, E.C. and Lobo, R.A. (1999) Do hyperandrogenic women with normal menses have polycystic ovarian syndrome? Fertil. Steril., 71, 319–322. Carter, G.D., Holland, S.M., Alaghband-Zadeh, J., Rayman, G., DorringtonWard, P. and Wise P.H. (1983) Investigation of hirsutism: testosterone is not enough. Ann. Clin. Biochem., 20, 262–263. Dabirashrafi, H. (1989) Complications of laparoscopic ovarian cauterization. Fertil. Steril., 52, 878–879. Eden, J.A., Place, J., Carter, G.D., Jones, J., Alaghband-Zadeh, J. and Pawson, M.E. (1989) The diagnosis of polycystic ovaries in subfertile women. Br. J. Obstet. Gynaecol., 96, 809–815. Felemban, A., Tan, S.L. and Tulandi, T. (2000) Laparoscopic treatment of polycystic ovaries with insulated needle cautery: a reappraisal. Fertil. Steril., 73, 266–269. Franks, S. (1995) Polycystic ovarian syndrome. N. Engl. J. Med., 333, 853–861. Gjonnaess, H. (1984) Polycystic ovarian syndrome treated by ovarian electrocautery through the laparoscope. Fertil. Steril., 41, 20–25. Goldzieher, J.W. and Axelrod, L.R. (1963) Clinical and biochemical features of polycystic ovarian disease. Fertil Steril., 14, 631–653. Judd, H.L., Rigg, L.A., Anderson, D.C. and Yen, S.S. (1976) The effects of ovarian wedge resection on circulating gonadotrophin and ovarian steroid levels in patients with polycystic ovary syndrome. J. Clin. Endocrinol. Metab., 43, 347–355. Li, T.C., Saravelos, H., Chow, M.S., Chisabingo, R. and Cooke, I.D. (1998) Factors affecting the outcome of laparoscopic ovarian drilling for polycystic ovarian syndrome in women with anovulatory infertility. Br. J. Obstet. Gynaecol., 105, 338–344. Loppo¨ hn, R.E. and Bogchelman, D.H. (1989) The relation of fertility and ovarian histology after bilateral ovarian wedge resection. Fertil. Steril., 52, 221–226. Merchant, R.N. (1996) Treatment of polycystic ovary disease with laparoscopic low-watt bipolar electrocoagulation of the ovaries. J. Am. Assoc. Gynecol. Laparosc., 3, 503–508. Naether, O.G.J., Baukloh, V., Fischer, R. and Kowalczyk, T. (1994) Longterm follow-up in 206 infertility patients with polycystic ovarian syndrome after laparoscopic electrocautery of the ovarian surface. Hum. Reprod., 9, 2342–2349. Naether, O.G.J., Fischer, R., Weise, H.C., Geiger-Kotzler, L., Delfs, T. and Rudolf, K. (1993) Laparoscopic electrocoagulation of the ovarian surface in infertile patients with polycystic ovarian disease. Fertil. Steril., 60, 88–94. Rossmanith, W.G., Keckstein, J., Spatzier, K. and Lauritzen, C. (1991) The impact of ovarian laser surgery on the gonadotrophin secretion in women with polycystic ovarian disease. Clin. Endocrinol. (Oxf.), 34, 223–230. Stein, I.F. (1964) Duration of fertility following ovarian wedge resection— Stein–Leventhal syndrome. West. J. Surg. Obstet. Gynecol., 78, 124–127. Tulandi, T. (1999) Laparoscopic treatment of polycystic ovarian syndrome. In Tulandi, T. (ed.) Atlas of Laparoscopic and Hysteroscopic Techniques for Gynecologists. W.B.Saunders, London, pp. 93–95. Weise, H.C., Naether, O., Fischer, R., Berger-Bispink, S. and Delfs, T. (1991) Behandlungsergebnisse mit der oberflachenkuaterisierung polyzystischer Ovarien bei Sterilitatspatientinnen. Geburtshilfe Frauenheilkunde, 51, 920–924. Submitted on August 13, 2001; accepted on November 19, 2001

1051