Presacral neurectomy for the treatment of pelvic pain associated with endometriosis: A controlled study Giovanni Battista Candiani, MD, Luigi Fedele, MD, Paolo Vercellini, MD, Stefano Bianchi, MD, and Giuliana Di Nola, MD Milan, Italy OBJECTIVE: Our objective was to evaluate the efficacy of presacral neurectomy combined with conservative surgery for the treatment of pelvic pain associated with endometriosis. STUDY DESIGN: In a randomized, controlled study performed in a tertiary institution 71 patients with moderate or severe endometriosis and midline dysmenorrhea were randomly assigned to conservative surgery alone (n = 36) or conservative surgery and presacral neurectomy (n = 35). Main outcome measures were relief of dysmenorrhea, pelvic pain, and deep dyspareunia after surgery according to a multidimensional and an analog pain scale. RESULTS: Presacral neurectomy markedly reduced the midline component of menstrual pain, but no statistically significant differences were observed between the two groups in the frequency and severity of dysmenorrhea, pelvic pain, and dyspareunia in the long-term follow-up. After presacral neurectomy, constipation developed or worsened in 13 patients and urinary urgency occurred in three and a painless first stage of labor in two. CONCLUSION: Presacral neurectomy should be combined with conservative surgery for endometriosis only in selected cases. (AM J OesTET GVNECOL 1992;167:100-3.)
Key words: Endometriosis, -presacral neurectomy, pelvic pain Presacral neurectomy first described by Jaboulayl and Ruggi2 in 1899, is widely used, in addition to removal of adhesions and ovarian and peritoneal endometriotic implants, in the treatment of pelvic pain associated with endometriosis. This surgical procedure is considered useful by many authors,'-6 although it is not without intraoperative risks and side effects. Moreover, almost all studies completed so far have been retrospective and have evaluated pain without an adequate investigation of the symptom itself or without the use of a pain scale. 7 In a recent prospective study Tjaden et al. 8 observed that midline dysmenorrhea was abolished in 100% of patients undergoing presacral neurectomy in addition to conservative surgery for endometriosis. However, their series was small, and the reported benefit exclusively concerned the midline component of dysmenorrhea, whereas no favorable effect was seen in the other components. We performed a prospective study of a larger series to evaluate the efficacy of presacral neurectomy in addition to conservative surgery in the relief of pelvic pain associated with moderate and severe endometriosis.
From the Department of Obstetrics and Gynecology, Universil:;' of Milan. ReceivedforpublicationAugust4, 1991; revised December 18,1991; accepted December 30, 1991_ Reprint requests: Luigi Fedele, MD, Department of Obstetrics and Gynecology, University of Milano, Via Commenda, 12,20122 Milano, Italy. 611 135912
100
Material and methods We recruited 78 consecutive patients with laparotomic or laparoscopic diagnosis of endometriosis stage III or IV of the revised American Fertility Society classification9 undergoing conservative surgery at the First Department of Obstetrics and Gynecology, University of Milan, from June 1986 to January 1990. All the women reported moderate or severe midline or midline plus lateral menstrual pelvic pain, associated in some with intermenstrual pain or deep dyspareunia. Thirty-four were infertile. None had previously undergone surgery for gynecologic disease or received medical treatment in the 6 months before the operation. We obtained informed consent from each patient and approval from the institution. The women were assigned to conservative surgery alone (n = 40) or to conservative surgery and presacral neurectomy (n = 38) according to a randomization list. Clinical and demographic data of the patients are shown in Table 1. The sample size was based on the assumption that dysmenorrhea would resolve in 50% of the women who did not undergo presacral neurectomy and in 80% to 90% of those who did. The number of patients included gave the study a power of 80% for a difference of 30% and a power of 90% for a difference of 40% at a level of statistical significance of 0.05. Before surgery all patients underwent gastroenterologic, urologic, and orthopedic evaluation to exclude other causes of pelvic pain. In addition, they all completed a questionnaire that requested information on
Presacral neurectomy for endometriosis
Volume 167 Number I
101
50 40 ~
:0 (1)
.0 0
....
a.
0~
r, -
, ----~--~-----~------:
r---~
30
;+ __ r---~-~~-~~----~--ll-------- ' J
20
J
10 2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
32
34
36
months Fig. 1. Probability of recurrence of moderate or severe dysmenorrhea, according to linear pain scale, after presacral neurectomy in association with conservative surgery (continuous line) and after conservative surgery alone (dotted line).
the presence, severity, and site of menstrual and pelvic pain and the presence and severity of deep dyspareunia. The severity of dysmenorrhea and pelvic pain was evaluated by means of a multidimensional scoring system, which is a modification of that proposed by Andersch and Milsom.lo This system defines pain according to limitation of working ability (unaffected = 0, rarely affected = 1, moderately affected = 2, clearly inhibited = 3), coexistence of systemic symptoms (absent = 0, present = 1), and need for analgesics (no = 0, rarely = I, regularly = 2, inefficacious = 3) and ranks their simple sum in three groups (0 to 3 = mild, 4 to 5 = moderate, 6 to 7 = severe). The women were also asked to specify the severity of dysmenorrhea and pelvic pain according to a 10-point linear pain scale in which 0 indicated the absence of pain, I to 5 mild pain, 6 to 7 moderate pain and 8 to 10 severe pain. The severity of deep dyspareunia was evaluated as described by Biberoglu and Behrman." The women were asked to record monthly any pain symptoms after surgery. Every 6 months they underwent a clinical examination and the pain diary was checked. The follow-up was carried out in all patients for at least 1 year. Conservative surgery was performed with the techniques described by Buttram and Reiter l2 and presacral neurectomy in accordance with Malinak's suggestions. 13 All adhesions and ovarian and peritoneal endometriotic implants were removed during surgery. The pattern of dysmenorrhea in the postoperative follow-up was analyzed with the life-table method." The patients with persistent or renewed, moderate, or severe pain were considered as treatment failures; the curve represents the cumulative probability rate of recurrence of moderate or severe dysmenorrhea. We used the t test for unpaired data to compare the means and the X2 test to analyze the frequencies.
Table I. Characteristics of patients
Age (yr, mean ± SD) No. of parous women No. of infertile women Endometriosis stage 1II
IV
Conservative surgery (n = 36)
Conservative surgery plus presacral neurectomy (n = 35)
31.1 ± 3.6 5
32.5 ± 4.2
19
15
26
23
lO
6
12
Results
Seven patients did not accept randomization, leaving 36 in the group for conservative surgery (group A) and 35 in that for conservative surgery plus presacral neurectomy (group B). According to the multidimensional and analog scales, dysmenorrhea before operation was moderate in 13 (36.2%) and eight (23%) of group A and in 10 (28.6%) and six (17.2%) of group Band severe in 23 (63.8%) and 28 (77%) of group A and in 25 (71.4%) and 29 (82.8%) of group B, respectively. In addition to midline dysmenorrhea, lateral menstrual pain was reported by 27 (75%) patients in group A and 24 (68.6%) in group B. The mean time (:!:: SD) taken to perform presacral neurectomy was 23 ± 8 minutes. No complications occurred during the operation, and no significant differences in postoperative morbidity (fever, days in lospital after surgery) were observed between the two groups. One patient underwent a second laparotomy 2 days after presacral neurectomy for a retroperitoneal presacral hematoma. Eleven subserous myomas (mean diameter 1.6 cm, range 0.6 to 3.5 cm) were enucleated or cauterized in eight patients of group A and 14 (mean diameter 1.7 cm, range 0.5 to 4 cm) in 10 patients of group B. Figs. 1 and 2 show the pattern of dysmenorrhea during follow-up in the two groups, according to the two
102 Candiani et al.
July 1992 Am J Obstet Gynecol
50 40
~
:0 !tJ
.0 0
...a.
0~
30 r--~.--H-1H----~~-H-------r++
20
r-----.H~r.---~
____~
r~----+~--+---'
10
• 2
4
8
6
10
12
14
16
18
20
22
24
26
28
30
32
34
36
months Fig. 2. Probability of recurrence of moderate or severe dysmenorrhea, according to multidimensional pain scale, after presacral neurectomy in association with conservative surgery (continuous line) and after conservative surgery alone (dotted line).
Table II. Frequency and severity of dysmenorrhea according to analog and multidimensional pain scale before and 12 months after surgery Conseroative surgery (n = 36) Dysmenorrhea
Before
I
After
Table III. Frequency and severity of pelvic pain and deep dyspareunia before and 12 months after surgery Conseroative surgery (n = 36)
Conservative surgery plus presacral neurectomy (n = 35) Before
I
After
Absent 23 (66%) 0 21 (58%) 0 Mild Analog 6 (17%) 0 0 504%) Multidimensional 0 7 (20%) 8 (22%) 0 Moderate Analog 8 (23%) 4 01 %) 6 (17%) 401%) Multidimensional 13 (36%) 4 (11 %) 10 (29%) 3 (9%) Severe Analog 28 (77%) 504%) 29 (83%) 2 (6%) Multidimensional 23 (64%) 3 (8%) 25 (71%) 2 (6%)
pain scales. Table II shows the frequency and severity of dysmenorrhea 12 months after surgery. In this period dysmenorrhea recurred in 15 patients in group A and in 12 in group B. Recurrent menstrual pain had a midline component in 15 (100%) group A women and in eight (66.7%) group B subjects (median = 3.525, P = 0.06) and a lateral one in nine (60%) and 10 (83%) (p = not significant), respectively. The frequency and severity of intermenstrual pain and deep dyspareunia in the two groups before and 12 months after surgery are reported in Table III. Presacral neurectomy was associated with different side effects in the follow-up. Moderate or severe constipation developed or worsened in 13 patients. One had recurrent subocelusive crises due to the formation of fecal masses in the rectal ampulla. Three patients reported the appearance of urinary urgency. Two women who deliv-
Before
Pelvic pain Absent Mild Analog Multidimensional Moderate Analog Multidimensional Severe Analog Multidimensional Deep dyspareunia Absent Mild Moderate Severe
I
After
Conservative surgery plus presacral neurectomy (n = 35)
Before
I
After
21 (58%) 31 (86%) 18 (51%) 32 (91%) 4 (11%) 2 (6%) 6 (17%) 3 (8%)
3 (9%) 0 6 (17%) 1 (3%)
6 (17%) 2 (6%) 7 (19%) 1 (3%)
8 (23%) 7 (20%)
5 (14%) 2 (6%)
607%) 2 (6%) 4 (11%) 1 (3%)
18 5 7 6
1 (3%) 1 (3%)
(50%) 29 (81%) (14%) 3 (8%) (19%) 1 (3%) (17%) 3 (8%)
16 4 6 9
(46%) (11%) (17%) (26%)
1 (3%) 1 (3%)
31 (89%) 0 2 (6%) 2 (6%)
ered after presacral neurectomy had a painless first stage of labor. One of them arrived at the hospital at 36 weeks of gestation, already in the second stage, with a breech presentation, and vaginal delivery was achieved without complications.
Comment In our study the addition of presacral neurectomy to conservative surgery for moderate or severe endometriosis did not result in a greater reduction of pelvic pain than did conservative surgery alone. In fact, 1 year after the operation 80% of the patients who underwent presacral neurectomy did not have moderate or severe dysmenorrhea, compared with 75% of the controls. Af-
Volume 167 Number I
ter conservative surgery plus presacral neurectomy deep dyspareunia resolved or improved in 84% of patients and intermenstrual pain in 82%, compared with 67% and 80% of women who underwent conservative surgery alone. Because of the relatively small size of our series, we cannot exclude the possibility that a small beneficial effect of presacral neurectomy on pain went unrecognized. However, our study had the power to demonstrate differences of 2:40%. Because presacral neurectomy may have important complications and side effects are frequent, an advantage of
Key words: Endometriosis, -presacral neurectomy, pelvic pain Presacral neurectomy first described by Jaboulayl and Ruggi2 in 1899, is widely used, in addition to removal of adhesions and ovarian and peritoneal endometriotic implants, in the treatment of pelvic pain associated with endometriosis. This surgical procedure is considered useful by many authors,'-6 although it is not without intraoperative risks and side effects. Moreover, almost all studies completed so far have been retrospective and have evaluated pain without an adequate investigation of the symptom itself or without the use of a pain scale. 7 In a recent prospective study Tjaden et al. 8 observed that midline dysmenorrhea was abolished in 100% of patients undergoing presacral neurectomy in addition to conservative surgery for endometriosis. However, their series was small, and the reported benefit exclusively concerned the midline component of dysmenorrhea, whereas no favorable effect was seen in the other components. We performed a prospective study of a larger series to evaluate the efficacy of presacral neurectomy in addition to conservative surgery in the relief of pelvic pain associated with moderate and severe endometriosis.
From the Department of Obstetrics and Gynecology, Universil:;' of Milan. ReceivedforpublicationAugust4, 1991; revised December 18,1991; accepted December 30, 1991_ Reprint requests: Luigi Fedele, MD, Department of Obstetrics and Gynecology, University of Milano, Via Commenda, 12,20122 Milano, Italy. 611 135912
100
Material and methods We recruited 78 consecutive patients with laparotomic or laparoscopic diagnosis of endometriosis stage III or IV of the revised American Fertility Society classification9 undergoing conservative surgery at the First Department of Obstetrics and Gynecology, University of Milan, from June 1986 to January 1990. All the women reported moderate or severe midline or midline plus lateral menstrual pelvic pain, associated in some with intermenstrual pain or deep dyspareunia. Thirty-four were infertile. None had previously undergone surgery for gynecologic disease or received medical treatment in the 6 months before the operation. We obtained informed consent from each patient and approval from the institution. The women were assigned to conservative surgery alone (n = 40) or to conservative surgery and presacral neurectomy (n = 38) according to a randomization list. Clinical and demographic data of the patients are shown in Table 1. The sample size was based on the assumption that dysmenorrhea would resolve in 50% of the women who did not undergo presacral neurectomy and in 80% to 90% of those who did. The number of patients included gave the study a power of 80% for a difference of 30% and a power of 90% for a difference of 40% at a level of statistical significance of 0.05. Before surgery all patients underwent gastroenterologic, urologic, and orthopedic evaluation to exclude other causes of pelvic pain. In addition, they all completed a questionnaire that requested information on
Presacral neurectomy for endometriosis
Volume 167 Number I
101
50 40 ~
:0 (1)
.0 0
....
a.
0~
r, -
, ----~--~-----~------:
r---~
30
;+ __ r---~-~~-~~----~--ll-------- ' J
20
J
10 2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
32
34
36
months Fig. 1. Probability of recurrence of moderate or severe dysmenorrhea, according to linear pain scale, after presacral neurectomy in association with conservative surgery (continuous line) and after conservative surgery alone (dotted line).
the presence, severity, and site of menstrual and pelvic pain and the presence and severity of deep dyspareunia. The severity of dysmenorrhea and pelvic pain was evaluated by means of a multidimensional scoring system, which is a modification of that proposed by Andersch and Milsom.lo This system defines pain according to limitation of working ability (unaffected = 0, rarely affected = 1, moderately affected = 2, clearly inhibited = 3), coexistence of systemic symptoms (absent = 0, present = 1), and need for analgesics (no = 0, rarely = I, regularly = 2, inefficacious = 3) and ranks their simple sum in three groups (0 to 3 = mild, 4 to 5 = moderate, 6 to 7 = severe). The women were also asked to specify the severity of dysmenorrhea and pelvic pain according to a 10-point linear pain scale in which 0 indicated the absence of pain, I to 5 mild pain, 6 to 7 moderate pain and 8 to 10 severe pain. The severity of deep dyspareunia was evaluated as described by Biberoglu and Behrman." The women were asked to record monthly any pain symptoms after surgery. Every 6 months they underwent a clinical examination and the pain diary was checked. The follow-up was carried out in all patients for at least 1 year. Conservative surgery was performed with the techniques described by Buttram and Reiter l2 and presacral neurectomy in accordance with Malinak's suggestions. 13 All adhesions and ovarian and peritoneal endometriotic implants were removed during surgery. The pattern of dysmenorrhea in the postoperative follow-up was analyzed with the life-table method." The patients with persistent or renewed, moderate, or severe pain were considered as treatment failures; the curve represents the cumulative probability rate of recurrence of moderate or severe dysmenorrhea. We used the t test for unpaired data to compare the means and the X2 test to analyze the frequencies.
Table I. Characteristics of patients
Age (yr, mean ± SD) No. of parous women No. of infertile women Endometriosis stage 1II
IV
Conservative surgery (n = 36)
Conservative surgery plus presacral neurectomy (n = 35)
31.1 ± 3.6 5
32.5 ± 4.2
19
15
26
23
lO
6
12
Results
Seven patients did not accept randomization, leaving 36 in the group for conservative surgery (group A) and 35 in that for conservative surgery plus presacral neurectomy (group B). According to the multidimensional and analog scales, dysmenorrhea before operation was moderate in 13 (36.2%) and eight (23%) of group A and in 10 (28.6%) and six (17.2%) of group Band severe in 23 (63.8%) and 28 (77%) of group A and in 25 (71.4%) and 29 (82.8%) of group B, respectively. In addition to midline dysmenorrhea, lateral menstrual pain was reported by 27 (75%) patients in group A and 24 (68.6%) in group B. The mean time (:!:: SD) taken to perform presacral neurectomy was 23 ± 8 minutes. No complications occurred during the operation, and no significant differences in postoperative morbidity (fever, days in lospital after surgery) were observed between the two groups. One patient underwent a second laparotomy 2 days after presacral neurectomy for a retroperitoneal presacral hematoma. Eleven subserous myomas (mean diameter 1.6 cm, range 0.6 to 3.5 cm) were enucleated or cauterized in eight patients of group A and 14 (mean diameter 1.7 cm, range 0.5 to 4 cm) in 10 patients of group B. Figs. 1 and 2 show the pattern of dysmenorrhea during follow-up in the two groups, according to the two
102 Candiani et al.
July 1992 Am J Obstet Gynecol
50 40
~
:0 !tJ
.0 0
...a.
0~
30 r--~.--H-1H----~~-H-------r++
20
r-----.H~r.---~
____~
r~----+~--+---'
10
• 2
4
8
6
10
12
14
16
18
20
22
24
26
28
30
32
34
36
months Fig. 2. Probability of recurrence of moderate or severe dysmenorrhea, according to multidimensional pain scale, after presacral neurectomy in association with conservative surgery (continuous line) and after conservative surgery alone (dotted line).
Table II. Frequency and severity of dysmenorrhea according to analog and multidimensional pain scale before and 12 months after surgery Conseroative surgery (n = 36) Dysmenorrhea
Before
I
After
Table III. Frequency and severity of pelvic pain and deep dyspareunia before and 12 months after surgery Conseroative surgery (n = 36)
Conservative surgery plus presacral neurectomy (n = 35) Before
I
After
Absent 23 (66%) 0 21 (58%) 0 Mild Analog 6 (17%) 0 0 504%) Multidimensional 0 7 (20%) 8 (22%) 0 Moderate Analog 8 (23%) 4 01 %) 6 (17%) 401%) Multidimensional 13 (36%) 4 (11 %) 10 (29%) 3 (9%) Severe Analog 28 (77%) 504%) 29 (83%) 2 (6%) Multidimensional 23 (64%) 3 (8%) 25 (71%) 2 (6%)
pain scales. Table II shows the frequency and severity of dysmenorrhea 12 months after surgery. In this period dysmenorrhea recurred in 15 patients in group A and in 12 in group B. Recurrent menstrual pain had a midline component in 15 (100%) group A women and in eight (66.7%) group B subjects (median = 3.525, P = 0.06) and a lateral one in nine (60%) and 10 (83%) (p = not significant), respectively. The frequency and severity of intermenstrual pain and deep dyspareunia in the two groups before and 12 months after surgery are reported in Table III. Presacral neurectomy was associated with different side effects in the follow-up. Moderate or severe constipation developed or worsened in 13 patients. One had recurrent subocelusive crises due to the formation of fecal masses in the rectal ampulla. Three patients reported the appearance of urinary urgency. Two women who deliv-
Before
Pelvic pain Absent Mild Analog Multidimensional Moderate Analog Multidimensional Severe Analog Multidimensional Deep dyspareunia Absent Mild Moderate Severe
I
After
Conservative surgery plus presacral neurectomy (n = 35)
Before
I
After
21 (58%) 31 (86%) 18 (51%) 32 (91%) 4 (11%) 2 (6%) 6 (17%) 3 (8%)
3 (9%) 0 6 (17%) 1 (3%)
6 (17%) 2 (6%) 7 (19%) 1 (3%)
8 (23%) 7 (20%)
5 (14%) 2 (6%)
607%) 2 (6%) 4 (11%) 1 (3%)
18 5 7 6
1 (3%) 1 (3%)
(50%) 29 (81%) (14%) 3 (8%) (19%) 1 (3%) (17%) 3 (8%)
16 4 6 9
(46%) (11%) (17%) (26%)
1 (3%) 1 (3%)
31 (89%) 0 2 (6%) 2 (6%)
ered after presacral neurectomy had a painless first stage of labor. One of them arrived at the hospital at 36 weeks of gestation, already in the second stage, with a breech presentation, and vaginal delivery was achieved without complications.
Comment In our study the addition of presacral neurectomy to conservative surgery for moderate or severe endometriosis did not result in a greater reduction of pelvic pain than did conservative surgery alone. In fact, 1 year after the operation 80% of the patients who underwent presacral neurectomy did not have moderate or severe dysmenorrhea, compared with 75% of the controls. Af-
Volume 167 Number I
ter conservative surgery plus presacral neurectomy deep dyspareunia resolved or improved in 84% of patients and intermenstrual pain in 82%, compared with 67% and 80% of women who underwent conservative surgery alone. Because of the relatively small size of our series, we cannot exclude the possibility that a small beneficial effect of presacral neurectomy on pain went unrecognized. However, our study had the power to demonstrate differences of 2:40%. Because presacral neurectomy may have important complications and side effects are frequent, an advantage of
