Influence of endometriosis on pain behaviors and ... - Semantic Scholar

Pain 95 (2002) 247–257 www.elsevier.com/locate/pain

Influence of endometriosis on pain behaviors and muscle hyperalgesia induced by a ureteral calculosis in female rats Maria Adele Giamberardino a,*, Karen J. Berkley b, Giannapia Affaitati a, Rosanna Lerza a, Lucia Centurione c, Domenico Lapenna a, Leonardo Vecchiet a a

Pathophysiology of Pain Laboratory, Department of Medicine and Science of Aging, ‘G. D’Annunzio’ University of Chieti, 66013 Chieti Scalo (CH),Italy b Program in Neuroscience, Florida State University, Tallahassee, FL 32306-1270, USA c Department of Biomorphology, ‘G. D’Annunzio’ University of Chieti, 66013 Chieti Scalo (CH), Italy Received 11 October 2000; received in revised form 30 July 2001; accepted 13 August 2001

Abstract Endometriosis and urinary calculosis can co-occur. Clinical studies have shown that both painful and non-painful endometriosis in women are associated with enhanced pain and referred muscle hyperalgesia from urinary calculosis, but the mechanisms underlying this phenomenon are still poorly understood. The aim of this study was to develop an animal model adequate to explore this viscero-visceral interaction in standardized conditions. Using a model of endometriosis previously developed to study reduced fertility and vaginal hyperalgesia, endometriosis (endo) or shamendometriosis (sham-endo) was induced in rats by autotransplantation of small pieces of uterus (or, for sham-endo, fat) on cascade mesenteric arteries, ovary, and abdominal wall. After the endometrial, but not the fat autografts had produced fluid-filled cysts (3 weeks), urinary calculosis was induced by implanting an artificial stone into one ureter. Pain behaviors were monitored by continuous 24-h videotape recordings before and after stone implantation. Referred muscle hyperalgesia was assessed by measuring vocalization thresholds to electrical stimulation of the oblique musculature (L1 dermatome). The data were compared with previously reported data from rats that had received only the stone. Neither endo nor sham-endo alone induced pain behaviors. Following stone implantation, in endo rats compared to sham-endo and stone-only rats, pain behaviors specifically associated with urinary calculosis were significantly increased and new pain behaviors specifically associated with uterine pathology became evident. Muscle hyperalgesia was also significantly increased. To explore the relationship between the amount of endometriosis and that of ureteral pain behavior, two separate groups of endo rats were treated with either a standard non-steroidal anti-inflammatory drugs (ketoprofen) or placebo from the 12th to the 18th day after endometriosis induction. The stone was implanted on the 21st day. Ketoprofen treatment compared to placebo significantly reduced the size of the cysts and both ureteral and uterine pain behaviors post-stone implantation. The size of the cysts showed a significant linear correlation with the poststone ureteral pain behaviors. In conclusion, endo increased pain crises and muscle hyperalgesia typically induced by a ureteral calculosis, and the ureteral calculosis revealed additional pain behaviors typically induced by uterine pathophysiology; and this enhancement was a function of the degree of endometriosis. This result closely reproduces the condition observed in humans and could be due to a phenomenon of ‘viscero-visceral’ hyperalgesia, in which increased input from the cyst implantation sites to common spinal cord segments (T10-L1) facilitates the central effect of input from the urinary tract. q 2002 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. Keywords: Uterus; Hysterectomy; Urinary tract; Pelvic pain; Colic; Ketoprofen

1. Introduction Endometriosis is a common clinical disorder, whose signs include the abnormal presence of viable endometrial tissue outside the uterus, usually in the abdominal/pelvic cavity, and whose main symptoms are reduced fertility, often severe dysmenorrhea (menstrual pain) and dyspareunia * Corresponding author. Tel./fax: 139-0871-565286. E-mail address: [email protected] (M.A. Giamberardino).

(vaginal hyperalgesia), and sometimes dyschezia (pain on defecation) (Wilson, 1987; Schenken, 1999; Farquhar, 2000; Frackiewicz, 2000; Woodward et al., 2001). A recent retrospective and prospective study in women found an interaction between dysmenorrhea, both primary and secondary to endometriosis, and urinary tract disturbances. Women with dysmenorrhea experienced a significantly higher number of urinary colic episodes from calculosis than non-dysmenorrheic women in a comparable period of time and the menstrual temporal patterns of the

0304-3959/02/$20.00 q 2002 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. PII: S 0304-395 9(01)00405-5

248

M.A. Giamberardino et al. / Pain 95 (2002) 247–257

episodes differed between the two groups (Giamberardino et al., 2001; see also Giamberardino, 2000). In addition, referred lumbar muscle hyperalgesia from urinary calculosis was significantly more intense in dysmenorrheic than in non-dysmenorrheic women, even when the number of colic episodes in the two groups was the same. A similar interaction was also found between latent algogenic conditions of the female reproductive organs, e.g. non-painful endometriosis (documented at laparoscopy) and urinary calculosis. Women with endometriosis and calculoses had significantly more colic episodes and referred lumbar muscle hyperalgesia than women with calculoses but without endometriosis in a comparable amount of time (Giamberardino et al., 2001). Although viscero-visceral interactions like the one just described are of considerable clinical importance for diagnosis and treatment, yet little is known of their mechanisms and etiology. Of value would be an animal model. Previous studies have established separate rat models of ureteral calculosis and endometriosis. Rats implanted with an artificial ureteral calculosis exhibit pain behaviors that mimic both the painful colic episodes and the referred muscle hyperalgesia experienced by humans (Vecchiet et al., 1989, 1992; Giamberardino et al., 1990, 1994, 1995). Similarly, rats in which endometriosis has been surgically induced exhibit some of the clinical symptoms reported by women; e.g. reduced fertility and vaginal hyperalgesia (Vernon and Wilson, 1985; Barragan et al., 1992; Dargenio et al., 1992; Berkley et al., 2001). In the present study, these two models were combined to test the hypothesis that endometriosis enhances pain behaviors and muscle hyperalgesia evoked by ureteral calculosis. Because non-steroidal antiinflammatory drugs (NSAIDs) have been shown to reduce pain associated both with endometriosis and ureteral calculosis (El-Baz and El-Tayeb Nasser, 1995; Ludwig, 1996; Shokeir et al., 1999) the influence of ketoprofen on the pain behaviors and muscle hyperalgesia was also evaluated. Some of the results have been published in abstract form (Giamberardino et al., 1998, 1999a,b).

2. Materials and methods A total of 56 female Sprague–Dawley rats (weighing 180–220 g) were used for the study. The study was approved by the Animal Care and Use Committee at the University of Chieti, Italy (protocol 19/00, revised 24/00) and adhered to the guidelines of the International Association for the Study of Pain (Zimmermann, 1983). Throughout the study, the rats were housed individually in a 35 £ 23 £ 18 cm plexiglass cage, with free access to food and water, maintained on a 12:12 h light/dark cycle, with artificial lighting from 08:00 to 20:00 h. To minimize distress, behaviors were assessed for the shortest amount of time necessary to obtain valid information (see below) before the rats were killed with an overdose of barbiturate. Although estrous stage was

monitored in most rats, it was not systematically included in the design of the study because previous studies found no significant influence of either estrous stage in which the stone is implanted or of the suppression of the estrus cycle via ovariectomy on pain behaviors subsequent to stone formation recorded over 4 days post-operatively (Giamberardino et al., 1997). 2.1. Rats with endometriosis/sham-endometriosis plus ureteral stone Thirty-six rats underwent surgical induction of either endometriosis (endo: n ¼ 18) or sham-endometriosis (sham-endo: n ¼ 18). Previous studies have shown that the endometrial autotransplants do not develop active cysts until about 2 weeks after the surgical induction (Vernon and Wilson, 1985). Therefore, on the 18th day after endo or sham surgery, ten each of endo and shamendo rats underwent implantation of stimulating wires into their left lateral oblique muscle. This muscle becomes hyperalgesic in rats implanted with an artificial stone in the left ureter (Giamberardino et al., 1990; 1995). Three days later (i.e. 21st day post-endo or post-sham-endo surgery), all rats in each group underwent stone implantation in the upper left ureter, at which time the endometrial or fat implants were examined to determine if cysts had formed and, if so, to measure their diameter. Because previous studies have shown that most of the pain behaviors associated with the stone occur over a period of about 4 days, rats were killed on the 4th day after stone implantation and an autopsy was done to evaluate the condition of the urinary tract. In a few endo rats (n ¼ 3), the cysts were removed to be processed histologically for light microscopy observation. Spontaneous pain behaviors were characterized and quantified by analyzing continuous (24 h) videotape recordings that were made for several days after endometriosis induction (16 rats; eight endo and eight sham-endo), and from 3 days pre- to 4 days post-stone implantation (all rats). Muscle sensitivity was assessed over the same time period in two ways: by examining whether or not the rats vocalized in response to gentle manual pressure on their oblique muscles and by quantitative measurements of vocalization thresholds to electrical stimulation of the left oblique muscle. 2.1.1. Surgical induction of endometriosis and shamendometriosis A previously established model of endometriosis that reduces fertility (Vernon and Wilson, 1985; Rajkumar et al., 1990; Barragan et al., 1992; Sharpe-Timms et al., 1998; Keenan et al., 1999; Schor et al., 1999) and produces vaginal hyperalgesia (Berkley et al., 2001) was used here. Rats were anesthetized with pentobarbital (50 mg/kg, intraperitoneally (i.p.)), the uterus was exposed through a midline abdominal incision, and a 1-cm segment of the


249

Fig. 1. Schematic representation of the surgical procedures for induction of endometriosis and ureteral stone. The cross-hatched squares indicate the five endometrial autografts (see text for details).

right uterine horn and associated fat tissue was removed. Five pieces of endometrium or, for the sham procedure, five pieces of fat were cut from this segment and were sewn around small vessels in various structures using nylon suture (Fig. 1) as follows: one on left ovary, three on alternate cascade mesenteric arteries that supply the caudal small intestine, one on the internal lower abdominal wall – right side at about the level of the L2 dermatome. 2.1.2. Ureteral stone implantation Using previously described procedures (Giamberardino et al., 1990, 1995), under pentobarbital anesthesia (50 mg/

kg, i.p.), the left ureter was approached via laparotomy and a 0.02 ml bolus of dental cement (lamell resin, Swedia) was injected, while still a fluid, into the upper third of the lumen, using a syringe with a 0.4 diameter needle (Fig. 1).

2.1.3. Muscle electrode implantation Under pentobarbital anesthesia (50 mg/kg, i.p.), insulated nickelchrome wires were implanted at the level of the left obliquus externus muscle. Connectors attached to the wires were fixed to the skull using techniques described in detail elsewhere (Giamberardino et al., 1990, 1995).

250


2.1.4. Quantification of spontaneous pain behaviors Spontaneous pain behaviors were analyzed from continuous videotape recordings by observers blind to the rat’s experimental group. Two types of pain behaviors were counted, ‘ureteral pain crises’ and ‘uterine pain behaviors’. As described previously (Giamberardino et al., 1995), ‘ureteral pain crises’ consisted of a sequence $three pain behaviors (of six possible) within a period of minimum 2 min duration. The six possible behaviors included: (i) ‘humpbacked’ position; (ii) ‘licking’ of the lower abdomen (and/ or the left flank); (iii) contraction of the left oblique musculature with inward moving of the ipsilateral hindlimb (‘inward’); (iv) stretching of the body with raised abdomen (‘stretch-stone’); (v) squashing of the lower abdomen against the floor (‘squash-stone’); and (vi) ‘supine’ position with left hindlimb adducted and compressed against the abdomen. ‘Uterine pain behaviors’ were those that occurred in the intervals between crises and consisted of four positions that have been described previously to occur specifically in response to inflammation of the uterus (Wesselmann et al., 1998) or distension (Berkley et al., 1995) and are easily distinguished from those that occur during ureteral crises. These four behaviors included: (i) ‘lambda’ position, when the rat suddenly hunches its back upwards into a sharp angle to form a triangular shape relative to the floor; (ii) ‘alpha’ position, with the abdomen adherent to the floor and the nose curving towards the tail on the affected side, forming a semicircle; (iii) ‘stretch-flat’ position, with stretching of the body with abdomen adherent to the floor; and (iv) ‘squash-pelvic’ position, with squashing of the lower part of the abdomen to the floor while in a standing or sitting position.

embedded in paraffin, and then 7 mm-thick sections were cut, placed on slides, stained with hematoxylin and eosin, and characterized light microscopically.

2.1.5. Evaluation of vocalization thresholds to electrical muscle stimulation Vocalization thresholds to electrical stimulation of the left obliquus externus muscle were measured daily at 09:30 h from 3 days prior through 4 days post-stone implantation using the method of limits. The stimuli were 250 ms trains of 1 ms square waves, frequency 200/s, delivered automatically every 2 s by a Lace Electronics constantcurrent stimulator. The intensity of the current was increased in 0.3 mA steps/2 s until vocalization occurred, then decreased in 0.1 mA steps until it disappeared and then increased again until vocalization returned, noting the corresponding values in mA. The threshold was calculated as the mean of these values (Giamberardino et al., 1990, 1995). Each day, the entire measurement procedure was repeated twice at 10-min intervals, with the second measure used for analysis.

2.3. Statistical analysis

2.1.6. Histology of cysts Cysts were removed and routinely processed for light microscopic evaluation, in the following manner: after immersion fixation in 10% formalin, the cysts were

2.2. Rats with endometriosis plus ureteral stone treated with ketoprofen This part of the study was designed to test the hypothesis that the amount of post-stone pain behavior is related to the extent of endometriotic lesions. Rats were therefore chronically treated with a conventional NSAID, ketoprofen, at the beginning of the process of cyst formation in an attempt to reduce their development, in comparison with placebo treatment. Twenty rats underwent endometriosis induction. They were then subdivided into two groups of ten each; rats of the first group were treated with ketoprofen (10 mg/kg/day, in two i.p. doses at 9:00 and 21:00 h) for a week, from the 12th to the 18th day after endometriosis induction (endo 1 keto) (Braga, 1990). Rats of the second group were treated with equivalent doses of placebo for the same period of time (endo 1 placebo). On the 21st day (after 3 days of washout), rats of both groups were implanted with a ureteral stone in the left ureter. In the course of this operation the number of cysts was counted and their diameter measured. All rats were videotaped for 3 days before and 4 days after stone implantation for recordings of ureteral and uterine pain behavior, with the same methodology as for the rats without treatment described in the previous experiment. They were sacrificed with an overdose of pentobarbital i.p., on the 4th day after stone implantation and submitted to autopsy for evaluation of the conditions of the urinary tract. Experimenters were unaware of the drug treatment condition throughout the study.

For each group of rats (endo, sham-endo, endo 1 keto, endo 1 placebo), means ^ SEM were calculated for the following parameters: (a) total duration of ‘ureteral crises’ during the 4-day post-stone implantation period, which included the sum of time spent in ureteral crises; (b) total duration of ‘uterine pain behaviors’ during the 4-day poststone implantation period, which included the sum of time spent in uterine behavior during the intervals between ureteral crises. For rats implanted with muscle electrodes, means ^ SEM were calculated of vocalization thresholds to electrical stimulation for each of the 3 days preceding and 4 days following stone implantation. For all endo rats (endo, endo 1 keto, and endo 1 placebo), means ^ SEM were also calculated of the number and diameter of cysts. The characteristics of the ‘ureteral crises’ and ‘uterine pain behaviors’ (total duration) obtained from the two groups of endo and sham-endo rats were compared with each other and with data previously obtained in a group of rats (stone-alone) that had undergone ureteral stone implantation (Giamberardino et al., 1995, 1997). The analysis was


performed via one-way analysis of variance (ANOVA) with post hoc t-tests. Vocalization thresholds to electrical stimulation for each group, endo, sham-endo, and stone-only rats were analyzed via ANOVA for repeated measures with post hoc t-tests to evaluate changes between pre- and post-stone implantation. For each group, the average post-stone decrease in threshold in the post-operative period was calculated (means ^ SEM). Data from the three groups were compared by means of one-way ANOVA with post hoc tests. The characteristics of the ‘ureteral crises’ and ‘uterine pain behaviors’ (total duration) and the number and diameter of the cysts obtained from the two groups of endo 1 keto and endo 1 placebo rats were compared with each other by means of Student’s t-test for unpaired samples. Correlation between the mean diameter of cysts and the amount of post-stone pain behavior was assessed using linear regression analysis. The level of significance was set at P , 0:05.

3. Results 3.1. Rats with endometriosis/sham-endometriosis plus ureteral stone 3.1.1. Spontaneous pain behaviors before stone implantation Rats exhibited no spontaneous pain behaviors after the induction of either endometriosis or sham-endometriosis and before stone implantation. Examination of the abdominal cavity at the time of stone implantation showed that the endometrial autografts had formed cysts in 15 of the 18 endometrium-grafted rats (almost all mesenteric and ovaric implants but not muscular implants), and that the fat autografts failed to form any cysts in any of the 18 fat-grafted rats. The mean number of cysts in the endo rats was 3:7 ^ 0:32 SEM, their mean length was 4:76 mm ^ 0:36 SEM. There was no significant difference in the weights of the rats in the different groups at this time or at the end of the study. 3.1.2. Spontaneous pain behaviors after stone implantation: ureteral crises The total duration of ureteral crises during the 4-day poststone implantation period was significantly longer in endo rats than in both stone-only and sham-endo rats. Surprisingly, however, the duration of ureteral crises was significantly shorter in the sham-endo rats than in the stone-only rats (Fig. 2). Given the limited number of rats in which the endometrial autografts had not produced cysts (n ¼ 3), the values of this group were not included in the statistical comparisons. However, endo rats with no cysts showed a mean total duration of crises very similar to that of shamendo rats (21:33 min ^ 13:04 SEM for endo-no cysts; 19:55 min ^ 7:11 SEM for sham-endo).

251

Fig. 2. Total duration of typical ureteral crises (sum of duration of all crises during the 4-day post-stone implantation period) in endo (n ¼ 15), stoneonly (n ¼ 40) and sham-endo rats (n ¼ 18). Means ^ SEM; one-way ANOVA (P , 0.0001) with post hoc tests: *, P , 0:05; **, P , 0:01; ***, P , 0:001.

3.1.3. Spontaneous pain behaviors after stone implantation: uterine pain behaviors The total duration of uterine pain behaviors during the 4day post-stone implantation period was significantly higher in the endo rats than in the sham-endo rats (Fig. 3). These behaviors were never observed in the stone-only rats. The mean total duration of these behaviors in the three endo rats with no cysts (46:39 min ^ 20:54 SEM) was shorter than

Fig. 3. Total time spent in uterine pain behavior (‘lambda’ 1 ‘alpha’ 1 ‘ stretch flat’ 1 ‘squash pelvic’ positions) between ureteral crises, in endo (n ¼ 15), stone-only (n ¼ 12) and sham-endo rats (n ¼ 18). Means ^ SEM; one-way ANOVA (P , 0:0001) with post hoc t-tests: ***, P , 0:001.

252


that of the endo rats with cysts (82:53 min ^ 12:45 SEM) but longer than that of the sham-endo rats (14:27 min ^ 2:66 SEM). 3.1.4. Muscle sensitivity After the induction of endometriosis or sham-endometriosis, but before stone implantation, none of the rats vocalized to gentle manual pressure on the oblique muscles. Basal mean vocalization thresholds to electrical stimulation of the oblique muscle did not differ between groups (endo, n ¼ 10, 3:68 mA ^ 0:62 SEM; stone-only, n ¼ 26, 3:34 mA ^ 0:33 SEM; sham-endo, n ¼ 10, 3:41 mA ^ 0:48 SEM). During the 4-day post-stone implantation period, all rats showed hyperalgesia of their left oblique muscle by vocalizing in response to gentle manual pressure on this muscle. However, the amount of hyperalgesia differed between groups in a pattern similar to the spontaneous pain behaviors (Figs. 4 and 5). In other words, the post-stone decrease in vocalization thresholds was significantly greater in endo than stone-only and sham-endo rats. Furthermore, shamendo rats had recovered from their hyperalgesia by the 4th day, whereas the endo and stone-only rats had not (Fig. 4).

Fig. 5. Average values of vocalization thresholds to electrical stimulation of the left obliquus externus muscle during 4 days post-stone implantation compared to pre-stone implantation values. Legend as for Fig. 3. Oneway ANOVA (P , 0:0002) with post hoc t-tests; ***, P , 0:001.

3.1.5. Histology of cysts The microscopic appearance of the endometrial cysts was the same regardless of where they were located, and was similar to that reported in earlier studies (Vernon and Wilson, 1985; Rajkumar et al., 1990). The implants were well vascularized and covered by a sheath of connective tissue. The cavity of the fluid-filled cysts contained numerous inflammatory cells, including polymorphonuclear cells, lymphocytes, plasmacells, and monocytes, with leukocytes being the most prevalent type (Fig. 6). The inner lining of the cyst consisted of simple columnar epithelium provided with a brush border. Stromal tissue was apparent, with very little glandular development, but marked vascularity and high inflammatory infiltration (Rajkumar et al., 1990). The cyst wall was thickest at the site of implantation.

Fig. 4. Vocalization thresholds to electrical stimulation of the left oblique musculature before (22, 2 1,0 days) and after (1–4 days) stone formation in the left ureter (performed on day 0) in rats studied up to 4 days postintervention (endo, n ¼ 10; stone-only, n ¼ 26, sham-endo, n ¼ 10). Thresholds are expressed as a percentage of basal values (mean of thresholds on days 22, 2 1,0). ANOVA for repeated measures to test the trend between pre- and post-intervention values. Endo group: P , 0:0002; P , 0:001 between basal values and 1st, 2nd, 3rd, and 4th day post-stone implantation; P , 0:001 between 1st and 2nd, 3rd and 4th day. Stoneonly group: P , 0:0002; P , 0:001 between basal values and 1st, 2nd, 3rd, and 4th day post-stone implantation. Sham-endo group: P , 0:0002; P , 0:01 between basal values and 1st and 3rd day, P , 0:001 between basal values and 2nd day after stone implantation.

3.1.6. Autopsy findings Two main patterns were observed in the urinary tract: (1) stone no longer detectable (presumably due to spontaneous elimination); (2) stone still present either in the implant position or displaced caudally within the ureter, with or without signs of spontaneous occlusion, i.e. dilated ureter proximal to the stone, increased kidney dimension. The incidence of pattern (1) was 12% for endo rats and 25% for sham-endo rats. In previous work on a sample of 21 stone-only rats (Giamberardino et al., 1995), the incidence of pattern (1) was 14%.


253

fen group were significantly less than those of the placebo group (Fig. 7A,B). A significant correlation was found between the mean diameter of cysts and the amount of ‘ureteral’ pain behavior recorded in all rats (P , 0:008; (r) 0.5787; Y ¼ 218:349 1 29:598 X). At autopsy, 10% of the rats in the placebo group and 20% in the ketoprofen group proved to have eliminated the stone. In the other rats, the stone remained in the lumen of the ureter, either in the original position, or displaced distally, with or without mild signs of occlusion.

4. Discussion

Fig. 6. Photomicrograph of an endometrial implant on the ovary, 25 days after implantation. Inflammatory cells (polymorphonuclear leukocytes, lymphocytes, and plasma cells) within the condensed material found in the lumen of the cyst; leukocytes were the most prevalent type (40 £ ). The inset shows the location of this condensed material in the lumen (right side of the inset; 10£).

3.2. Rats with endometriosis plus ureteral stone treated with ketoprofen Endo animals treated with placebo all developed cysts (in the mesentery and on the ovary) whose mean number and mean diameter did not differ significantly from those recorded in endo rats with no treatment who developed stones (reported in the first part of the study). Their poststone ‘ureteral’ and ‘uterine’ pain behaviors were also not significantly different from those recorded in endo rats with no treatment. Endo animals treated with ketoprofen all developed cysts similar in number to those in the placebo group (respectively, 3:3 ^ 0:21 and 3:5 ^ 0:34 SEM), but their mean diameter was reduced of about 50%, and the difference was statistically significant (Fig. 7C). The poststone ‘ureteral’ and ‘uterine’ pain behaviors in the ketopro-

In female rats subjected to experimental endometriosis, the endometrial autografts that had previously been shown to produce vaginal hyperalgesia (Berkley et al., 2001) did not by themselves produce spontaneous pain behaviors or muscle hyperalgesia. The procedure, however, had profound influences on pain behaviors and referred muscle hyperalgesia produced by the artificial ureteral stone. The endometrial autografts increased pain behaviors and exacerbated referred muscle hyperalgesia associated with the ureteral stone. Furthermore, ureteral stone implantation uncovered additional pain behaviors associated with uterine pathophysiology. In addition, the increased post-stone pain behavior was influenced by the degree of endometrial development, because the amount of pain behavior correlated linearly with the size of the cysts. In other words, partial hysterectomy and the implantation of cyst-producing endometrial autografts created a latent pelvic algogenic condition which augmented and was augmented by pathophysiology of the ureter. In rats subjected to sham experimental endometriosis, the implantation of fat autografts per se did not produce any change in behavior. In contrast to the endo procedure, however, the sham endometrial surgery produced significant changes in the post-stone behavior, which, unexpectedly,

Fig. 7. (A) Total duration of typical ureteral crises (sum of duration of all crises during the 4-day post-stone implantation period); (B) total time spent in uterine pain behavior (‘lambda’ 1 ‘alpha’ 1 ‘stretch flat’ 1 ‘squash pelvic’ positions) between ureteral crises; (C) diameter of endometriosis cysts (21 days after surgical induction of endometriosis) in two groups of ten endo rats each, treated with placebo (placebo) and 10 mg/kg/day of ketoprofen (ketoprofen) from the 12th to the 18th day post-endometriosis induction, respectively. Stone formation performed on the 21st day after endometriosis. ***, P , 0:001; Student’s ttest for unpaired samples.

254


were opposite to those produced by the endo procedure, i.e. decreased spontaneous ureteral pain crises and reduced duration of referred muscle hyperalgesia. The rats with sham endometrial surgery also exhibited fewer uterine pain behaviors than those observed in the rats with cystforming grafts. In other words, partial hysterectomy combined with implantation of non-cyst-forming autografts of fat tissue not only failed to give rise to a latent algogenic condition, but also reduced pain behaviors and muscle hyperalgesia associated with pathophysiology of the ureter. These opposite results obtained by combining either endometriosis with ureteral stone or sham-endometriosis with ureteral stone were surprising. Because the two autograft surgical procedures were similarly invasive, it had been anticipated that if the sham surgery was to have any effect it would be in the same direction as, but perhaps less than any effect produced by the endometrial surgery. 4.1. Possible mechanisms The main difference between the two procedures was that the endometrial but not the fat autografts developed into cysts. These cysts have been shown to secrete, and release into the abdomen, endometrial and other substances, including algogenic prostaglandins (Vernon and Wilson, 1985; Dargenio et al., 1992; Sharpe and Vernon, 1993). They also cause an increased concentration of interleukin-6 in peripheral serum (Lim and Schenken, 1993). No such substances are found following the fat implants. 4.1.1. Endometrial autotransplants Given the effects of the cysts, the failure of endometrial autografts to induce spontaneous pain behaviors would at first seem surprising. It should be remembered, however, that more than 20% of women with documented endometriosis do not experience spontaneous pelvic pain; most of the reported pains are induced during menstruation, by defecation, or by coitus (Farquhar, 2000). Furthermore, the grafts do produce vaginal hyperalgesia in rats (Berkley et al., 2001), like they do in women with endometriosis (Ripps and Martin, 1993; Farquhar, 2000; Frackiewicz, 2000). In the experiments here, they also produced a latent algogenic condition in which pain behaviors and referred muscle hyperalgesia induced by a ureteral calculosis were augmented. These results may thus be the experimental counterpart of the clinical condition in which pain crises associated with urinary stones are increased in women who suffer from a latent algogenic condition of the female reproductive organs, like asymptomatic endometriosis (Giamberardino et al., 2001). The neural mechanisms underlying this mutual exacerbation are unknown but a reasonable hypothesis is that an increased input to the nervous system from one visceral domain sensitizes convergent viscero-viscero-somatic sensory neurons so that the effects of inputs from other visceral domains are enhanced (Giamberardino, 2000; see

also Cervero, 1995). Such effects could involve interactions in any one of or cooperation between at least three regions: peripheral autonomic ganglia, solitary nucleus, and spinal cord. The ureter and uterine horn are both innervated by afferents in the hypogastric nerve that travel via the inferior mesenteric ganglion (IMG) to lower thoracic/upper lumbar segments in the spinal cord (Peters et al., 1987; Marfurt and Echtenkamp, 1991; Semenenko and Cervero, 1992; Papka and Traurig, 1993; Berkley et al., 1993b). Some uterine and ureteral afferents also travel in the pelvic nerve via the pelvic ganglion (PG) to the lower lumbar/upper sacral segments (Saria et al., 1983; Papka and Traurig, 1993; Berkley et al., 1993b; Collins et al., 1999). Regions where the grafts produced cysts (ovary and mesenteric arterial cascade vessels) are innervated by splanchnic nerve fibers traveling via the superior mesenteric and coeliac ganglia to lower thoracic/upper lumbar segments as well as via the IMG (Ja¨ nig and Morrison, 1986; Papka and Traurig, 1993). It is likely that many of these afferents were sensitized by the prostaglandins and cytokines produced by the cysts and surgical trauma (Levine and Reichling, 1999). The fact that the degree of increased post-stone pain behavior correlated directly with the dimensions of the cysts, and, presumably, the amount of inflammatory products, seems in line with this hypothesis. Because branches of the afferents form synapses on postganglionic neurons within both the IMG and PG (King and Szurszewski, 1989; Papka, 1990; Collins et al., 2000), some of the uterus–ureter interactions observed here could have occurred within the peripheral autonomic ganglia. Such interactions within the ganglia, however, do not explain the effects of the endometrial transplants on muscle sensitivity (i.e. viscero-somatic interactions). These actions could have occurred within the caudal spinal cord where viscero-viscero-somatic convergence between reproductive, gastrointestinal, urinary, and caudal somatic regions is well documented and has been hypothesized to serve as a substrate for referred hyperalgesia in muscle and other viscera (Berkley et al., 1993a; Giamberardino, 2000). It is also possible that neural events within solitary nucleus are somehow involved. In addition to the peripheral sensory innervation described above, all regions, including the ureter and uterus, are innervated by vagal afferent fibers that convey information to the solitary nucleus of the brainstem (Ortega-Villalobos et al., 1990; Hubscher and Berkley, 1995; Komisaruk et al., 1996; Collins et al., 1999). Furthermore, neurons there respond convergently to stimulation of both reproductive and gastrointestinal organs (Hubscher and Berkley, 1994). Vagal activation is usually associated with inhibition of nociception and nociceptive behavior, acting indirectly on spinal neurons by way of the solitary nucleus and its output pathways (Fu et al., 1992; Ja¨ nig et al., 2000). However, activation of myelinated vagal afferents has been shown to facilitate somatic cutaneous nociception (Randich and Gebhart, 1992). Thus, one possibility here is that vagal


activity that was produced by the abdominal cysts influenced, via a brainstem loop, either the activity of the afferents that supply the stone-implanted ureter (e.g. see Cunningham and Sawchenko, 1990) or their actions within the spinal cord (Foreman, 2000; Ja¨ nig et al., 2000). 4.1.2. Endometrial autotransplants and ketoprofen Chronic treatment of the endo rats with a standard NSAID (ketoprofen) that started at the time cysts began to develop resulted in a diminished formation of endometriotic lesions (cysts of smaller dimensions, though of a similar number) and in a significant reduction in post-stone pain behavior, in comparison with chronic treatment of endo rats with placebo. This result demonstrates the interdependence between the cysts (their dimensions and thus presumably their degree of activity) and the changes in pain behavior recorded post-operatively. Although NSAIDs are not the primary choice in the treatment of endometriosis in the clinical setting, they are often used to help keep the symptoms of this disease under control (Petruzzelli et al., 2001), and have been shown to have a cytostatic effect on the lesions, i.e. by helping to prevent their further growth. While NSAID treatment in the rats here appears more effective than it is in patients, this difference may be due to the different treatment protocols. The rats began their treatment when the cysts were forming, thus preventing in part complete cystic formation, while women normally begin treatment either when symptoms become manifest or when the lesions are already evident at laparoscopy. 4.1.3. Sham endometrial autotransplants While there are many plausible mechanisms that could underlie the mutual augmentation of pain behaviors and referred hyperalgesia produced by the inflammatory endometrial autotransplants, it is difficult to envision mechanisms underlying the surprising reduction in ureteral pain behaviors and muscle hyperalgesia produced by the sham endometrial surgical procedures. Examination of the regions where the fat autotransplants were sewn revealed the presence of only the sutures, with no signs of any remaining fat or inflammation around them and no cyst formation. Furthermore, there were no differences in the body weights of the two groups. Thus, there was something about the surgical manipulations that gave rise to the reduction in ureteral pain behaviors and muscle hyperalgesia, e.g. partial hysterectomy, brief exposure to the fat implants, or sutures. Considering that partial hysterectomy was performed in both the endo and sham endo groups, it is unlikely for it alone to be responsible for the different outcome in behavior subsequent to stone implantation in the two groups; it is therefore more probable that some variable related to the surgical procedure at the site of the implants (of endo tissue or of fat) are involved to the hypoalgesic effect of the sham procedure. Another possible contributing, but not completely explanatory factor, is that about twice as many sham-endo rats (25%) as compared with endo

255

rats (12%) and stone-only rats (14%) were found to have lost their stones at the time of autopsy, suggesting increased ureter contractility in the sham-endo rats. Stone elimination is, in fact, known to be associated in patients with at least a reduction of spontaneous pain from the urinary tract (see Vecchiet et al., 1989, 1992; Giamberardino et al., 1994). Studies to isolate which of these factors produced the hypoalgesic effects in sham-operated rats are underway. Meanwhile, it may be useful to consider the relative activation of the various afferents in the two experimental procedures. For the endometrial transplants, the cysts created a chronic algogenic activation of vagal and splanchnic afferent fibers from a number of abdominal regions that would have been added to the activation of a different set of vagal and pelvic/hypogastric afferents by the partial hysterectomy and ureteral stone. Thus, the added ‘algogenic load’ for the endometrial implants was more widespread and intense than that from the fat transplants. One highly speculative but testable possibility then is that the main difference between the two procedures is differential engagement of the vagal regulatory system so that in the case of the endometrial implants, but not the fat implants, vagal inhibition arising from the hysterectomy and implant sutures (Randich and Gebhart, 1992; Ja¨ nig et al., 2000) would be overcome. Clinical counterparts to the reduction in stone-induced pain behaviors are not documented. However, it is well known clinically that not all individuals who pass ureteral stones suffer colic episodes, referred pain, or referred hyperalgesia (Giamberardino et al., 1994). Despite much speculation, little is in fact known about the mechanisms that underlie this variance. This failure of ureteral stone passage to produce pain may involve mechanisms similar to those that underlie the failure of cardiac ischema to produce pain (‘silent ischemia’), but again this phenomenon is not well understood (Sylven, 1997; Krone, 1998). The results here suggest that part of the explanation for silent ischemia as well as ‘silent calculosis’ might lie in the state of health of other internal organs. In conclusion, the results of the present study provide clear experimental evidence of phenomena of viscero-visceral interactions with significant clinical relevance for both diagnosis and therapy. Further studies on this animal model of combined abdominal–pelvic and ureteral pathophysiology will allow further exploration of the mechanisms of these interactions that could lead to the development of more effective diagnostic and therapeutic strategies.

Acknowledgements This study was supported by University Research Funds from Chieti, Italy, NIH grant RO-1 NS11892, and NATO Collaborative Research Grant (SA5-05 CRG 971080). G.A. was supported by a grant from the Istituto UPSA per il

256


Dolore (Italy). We thank Eric Carr for help with Fig. 1, and Dr Robert Foreman for constructive discussions.

References Barragan JC, Brotons J, Ruiz JA, Acien P. Experimentally induced endometriosis in rats: effect on fertility and the effects of pregnancy and lactation on the ectopic endometrial tissue. Fertil Steril 1992;58:1215– 1219. Berkley KJ, Hubscher CH, Wall PD. Neuronal responses to stimulation of the cervix, uterus, colon and skin in the rat spinal cord. J Neurophysiol 1993a;69:545–556. Berkley KJ, Robbins A, Sato Y. Functional differences between afferent fibers in the hypogastric and pelvic nerves innervating female reproductive organs in the rat. J Neurophysiol 1993b;69:533–544. Berkley KJ, Wood E, Scofield SL, Little M. Behavioral responses to uterine or vaginal distension in the rat. Pain 1995;61:121–131. Berkley KJ, Cason A, Jacobs H, Bradshaw HB, Wood E. Vaginal hyperalgesia in a rat model of endometriosis. Neurosci Lett 2001;306:185– 188. Braga PC. Ketoprofen: i.c.v. injection and electrophysiological aspects of antinociceptive effect. Eur J Pharmacol 1990;184:273–280. Cervero F. Visceral pain: mechanisms of peripheral and central sensitization. Ann Med 1995;27:235–239 Special section: modern trends in postoperative pain control. Collins JJ, Lin CE, Berthoud HR, Papka RE. Vagal afferents from the uterus and cervix provide direct connections to the brainstem. Cell Tissue Res 1999;295:43–54. Collins JJ, Wilson K, Fischer-Colbrie R, Papka RE. Distribution and origin of secretoneurin-immunoreactive nerve in the female rat uterus. Neuroscience 2000;95:255–264. Cunningham Jr ET, Sawchenko PE. Central neural control of esophageal motility: a review. Dysphagia 1990;5:35–51. Dargenio R, Corbucci MG, Lamanna MA, Garcea N. Indomethacin and fertility in experimental endometriosis. Acta Eur Fertil 1992;23:85–88. El-Baz MA, El-Tayeb Nasser M. Emergency treatment of renal colic with intravenous ketoprofen. Int Urol Nephrol 1995;27:245–249. Farquhar CM. Extracts from the ‘clinical evidence’. Endometriosis. BMJ 2000;320:1449–1452. Frackiewicz EJ. Endometriosis: an overview of the disease and its treatment. J Am Pharm Assoc (Wash) 2000;40:645–657. Foreman RD. Integration of viscerosomatic sensory input at the spinal level. In: Mayer EA, Saper CB, editors. Progress in Brain Research, vol. 122. Amsterdam: Elsevier, 2000. pp. 209–221. Fu Q-G, Chandler MJ, McNeill DL, Foreman RD. Vagal afferent fibers excite upper cervical neurons and inhibit activity of lumbar spinal neurons in the rat. Pain 1992;51:91–100. Giamberardino MA. Visceral hyperalgesia. In: Devor M, Rowbotham MC, Wiesenfeld Halin Z, editors. Progress in Pain Research and Management, Seattle, WA: IASP Press, 2000. pp. 523–550. Giamberardino MA, Vecchiet L, Albe-Fessard D. Comparison of the effects of ureteral calculosis and occlusion on muscular sensitivity to painful stimulation in rats. Pain 1990;43:227–234. Giamberardino MA, de Bigontina P, Martegiani C, Vecchiet L. Effects of extracorporeal shock-wave lithotripsy on referred hyperalgesia from renal/ureteral calculosis. Pain 1994;56:77–83. Giamberardino MA, Valente R, de Bigontina P, Vecchiet L. Artificial ureteral calculosis in rats: behavioural characterization of visceral pain episodes and their relationship with referred lumbar muscle hyperalgesia. Pain 1995;61:459–469. Giamberardino MA, Affaitati G, Valente R, Iezzi S, Vecchiet L. Changes in visceral pain reactivity as a function of estrous cycle in female rats with artificial ureteral calculosis. Brain Res 1997;744:234–238. Giamberardino MA, Affaitati G, Vecchiet L, Berkley KJ. Effects of endo-

metriosis on pain behaviours induced by ureteral calculosis in female rats. J Musculoske Pain 1998;6:172. Giamberardino MA, Affaitati G, Lerza R, Vecchiet L, Berkley J. The impact of painful gynecological conditions on pain of urological origin. Soc Neurosci Abstr 1999a;25:143. Giamberardino MA, Berkley KJ, Affaitati G, Lerza R, Vecchiet L. The influence of endometriosis on pain behaviors induced by ureteral calculosis in female rats. IASP Abstracts, Seattle, WA: IASP Press, 1999b p. 392. Giamberardino MA, De Laurentis S, Affaitati G, Lerza R, Lapenna D, Vecchiet L. Modulation of pain and hyperalgesia from the urinary tract by algogenic conditions of the reproductive organs in women. Neurosci Lett 2001;304:61–64. Hubscher CH, Berkley KJ. Responses of neurons in caudal solitary nucleus of female rats to stimulation of vagina, cervix, uterine horn and colon. Brain Res 1994;664:1–8. Hubscher CH, Berkley KJ. Spinal and vagal influences on the responses of rat solitary nucleus neurons to stimulation of uterus, cervix and vagina. Brain Res 1995;702:251–254. Ja¨ nig W, Morrison JFB. Functional properties of spinal visceral afferents supplying abdominal and pelvic organs, with special emphasis on visceral nociception. In: Cervero F, Morrison JFB, editors. Progress in Brain Research, vol. 67. Amsterdam: Elsevier, 1986. pp. 87–114. Ja¨ nig W, Khasar SG, Levine JD, Miao FJ-P. The role of vagal visceral afferents in the control of nociception. In: Mayer EA, Saper CB, editors. Progress in Brain Research, vol. 122. Amsterdam: Elsevier, 2000. pp. 273–287. Keenan JA, Williams-Boyce PK, Massey PJ, Chen TT, Caudle MR, Bukovsky A. Regression of endometrial explants in a rat model of endometriosis treated with the immune modulators loxoribine and levamisole. Fertil Steril 1999;72:135–141. King BF, Szurszewski JH. Peripheral reflex pathways involving abdominal viscera: transmission of impulses through prevertebral ganglia. Am J Physiol 1989;256:581–588. Komisaruk BR, Bianca R, Sansone G, Gomez LE, Cueva-Rolon R, Beyer C, Whipple B. Brain-mediated responses to vaginocervical stimulation in spinal cord-transected rats: role of the vagus nerves. Brain Res 1996;708:128–134. Krone RJ. Diagnosis and prognosis of silent ischemia. Cardiologia 1998;43:1159–1168. Levine JD, Reichling DB. Peripheral mechanisms of inflammatory pain. In: Wall PD, Melzack R, editors. Textbook of Pain, 4th ed.. Edinburgh: Churchill Livingstone, 1999. pp. 59–84. Lim YT, Schenken RS. Interleukin-6 in experimental endometriosis. Fertil Steril 1993;59:912–916. Ludwig H. Dysmenorrhea. Ther Umsch 1996;53:431–441. Marfurt CF, Echtenkamp SF. Sensory innervation of the rat kidney and ureter as revealed by the anterograde transport of wheat germ agglutinin-horseradish peroxidase (WGA-HRP) from dorsal root ganglia. J Comp Neurol 1991;311:389–404. Ortega-Villalobos M, Garcia-Bazan M, Solano-Flores LP, Ninomiya-Alarcon JG, Guevara-Guzman R, Wayner MJ. Vagus nerve afferent and efferent innervation of the rat uterus: an electrophysiological and HRP study. Brain Res Bull 1990;25:365–371. Papka RE. Some nerve endings in the rat pelvic paracervical autonomic ganglia and varicosities in the uterus contain calcitonin gene-related peptide and originate from dorsal root ganglia. Neuroscience 1990;39:459–470. Papka RE, Traurig HH. Autonomic efferent and visceral sensory innervation of the female reproductive system: special reference to neurochemical markers in nerves and ganglionic connections. In: Maggi CA, editor. Nervous Control of the Urogenital System, Chur, Switzerland: Harwood Academic, 1993. pp. 421–464. Peters LC, Kristal MB, Komisaruk BR. Sensory innervation of the external and internal genitalia of the female rat. Brain Res 1987;408:199–204. Petruzzelli P, Montironi PL, Fedele M, Di Noto, C. Dolore pelvico da endometriosi. In: Margaria E, Gollo E, editors. Proceedings of the

M.A. Giamberardino et al. / Pain 95 (2002) 247–257 XXIII Congress of the Italian Chapter of IASP. Milan: Springer, 2001. p. 211–218. Rajkumar K, Schott PW, Simpson CW. The rat as an animal model for endometriosis to examine the recurrence of ectopic endometrial tissue after regression. Fertil Steril 1990;53:921–925. Randich A, Gebhart GF. Vagal afferent modulation of nociception. Brain Res Brain Res Rev 1992;17:77–99. Ripps BA, Martin DC. Endometriosis and chronic pelvic pain. Obstet Gynecol Clin North Am 1993;20:709–717. Saria A, Lundberg JM, Hua X, Lembeck F. Capsaicin-induced substance P release and sensory control of vascular permeability in the guinea-pig ureter. Neurosci Lett 1983;41:167–172. Schenken RS. In: Scott JR, Di Saia PJ, Hammond CB, Spellacy WN, editors. Danforth’s Obstetrics and Gynecology, 8th ed.. Philadelphia, PA: Lippincott Williams and Wilkins, 1999. pp. 669–675. Schor E, Baracat EC, Simoes MJ, de Freitas V, Giannotti Filho O, de Lima GR. Effects of conjugated estrogens and progestogen in surgically induced endometriosis in oophorectomized rats. Clin Exp Obstet Gynecol 1999;26:158–161. Semenenko FM, Cervero F. Afferent fibers from the guinea-pig ureter: size and peptide content of the dorsal root ganglion cells of origin. Neuroscience 1992;47:197–201. Sharpe KL, Vernon MW. Polypeptides synthesized and released by rat ectopic uterine implants differ from those of the uterus in culture. Biol Reprod 1993;48:1334–1340. Sharpe-Timms KL, Zimmer RL, Jolliff WJ, Wright JA, Nothnick WB,

257

Curry TE. Gonadotropin-releasing hormone agonist (GnRH-a) therapy alters activity of plasminogen activators, matrix metalloproteinases, and their inhibitors in rat models for adhesion formation and endometriosis: potential GnRH-a-regulated mechanisms reducing adhesion formation. Fertil Steril 1998;69:916–923. Shokeir AA, Abdulmaabound M, Farage Y, Mutabagani H. Resistive index in renal colic: the effect of nonsteroidal anti-inflammatory drugs. BJU Int 1999;84:249–251. Sylven C. Neurophysiological aspects of angina pectoris. Z Kardiol 1997;86(1):95–105. Vecchiet L, Giamberardino MA, Dragani L, Albe-Fessard D. Pain from renal/ureteral calculosis: evaluation of sensory thresholds in the lumbar area. Pain 1989;36:289–295. Vecchiet L, Giamberardino MA, de Bigontina P. Referred pain from viscera: when the symptom persists despite the extinction of the visceral focus. Adv Pain Res Ther 1992;20:101–110. Vernon MW, Wilson EA. Studies on the surgical induction of endometriosis in the rat. Fertil Steril 1985;44:684–694. Wesselmann U, Czakanski PP, Affaitati G, Giamberardino MA. Uterine inflammation as a noxious visceral stimulus: behavioral characterization in the rat. Neurosci Lett 1998;246:73–76. Wilson EA, editor. Endometriosis New York, NY: Alan R. Liss, 1987. Woodward P, Sohaey R, Mezzetti Jr T. Endometriosis: radiologic-pathologic correlation. Radiographics 2001;21:193–216. Zimmermann M. Ethical guidelines for investigation of experimental pain in conscious animals. Pain 1983;16:109–110.