| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Genitourinary Imaging |
1 From the Department of Radiology, Dartmouth Medical School, Dartmouth-Hitchcock Medical Center, One Medical Center Dr, Lebanon, NH 03756 (R.D.H., S.R.H.), and Dartmouth College, Hanover, NH (A.P.). Received June 1, 1999; revision requested July 26; revision received November 1; accepted November 24. A.P. supported in part by a grant from the Dartmouth Women in Science Project, Dartmouth College. Address correspondence to R.D.H. (e-mail: robert.harris@hitchcock.org).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Data from 86 female patients with pelvic pain and normal pelvic US findings seen in a US section over a 15-month period were evaluated 6–21 months after US. Medical chart review follow-up was available in 86 patients, and telephone interview follow-up was conducted in 85 patients. We collected data on the outcome of pain; subsequent imaging, treatment, and surgery; and the duration of pain before US.
RESULTS: Pelvic pain improved or resolved in 66 (77%) of the 86 patients. In the group with acute or subacute pain (duration 
6 months), 62 (86%) of the 72 patients (19 with acute pain and 53 with subacute pain) had improvement or resolution of symptoms. In the group with chronic pain (duration > 6 months), seven (50%) of the 14 patients had improved symptoms. Further imaging (13 studies) was performed in nine patients: Twelve studies were normal, and one computed tomographic scan (1 month after the first US examination) showed diverticulitis. Eleven patients underwent 19 surgical procedures (endometrial sampling, hysteroscopy, laparoscopy, or hysterectomy). Four demonstrated clinically important disease (endometriosis and pelvic varices, endometriosis, adenomyosis, or pelvic adhesions).
CONCLUSION: The majority of patients with pelvic pain and normal pelvic US findings had improvement or resolution of their symptoms, and those with acute or subacute pain were more likely to report improvement or resolution of pain than those with chronic pain. The yield of further imaging studies was low, and disease was identified in a minority of patients.
Index terms: Endometriosis, 85.3192 • Pelvic organs, abnormalities, 85.31, 85.3192, 85.80 • Pelvic organs, diseases, 85.31, 85.3192, 85.80 • Pelvic organs, US, 85.1298 • Radiology and radiologists, outcomes studies • Ultrasound (US), utilization
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
US is generally accepted as the first imaging modality used in patients with pelvic pain. Patients with pelvic pain compose a substantial proportion of the referrals to our US department. In our experience, a substantial number of patients with pelvic pain will have normal US findings. While much has been published that correlates positive US findings with surgical findings (6–10), little is known about the outcomes in those patients with negative US findings, other than the results of a study by Barloon et al (11), who found a high negative predictive value (92% [54 of 59]) for normal endovaginal US findings in a myriad of clinical settings. Our study was designed to examine the clinical outcome of patients with pelvic pain who had normal pelvic US findings.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Eighty-six of these patients had follow-up available and were included in the study; 14 patients were lost to follow-up. Medical records were reviewed (together by R.D.H. and A.M.P.) in all cases, and telephone interviews were conducted (by R.D.H. or A.M.P.) with 85 patients (one patient had a nonworking phone number). A normal pelvic US finding was defined as adequate US depiction of a normal uterus and endometrium for the patient’s menstrual and parous status. In addition, both ovaries were well depicted and unremarkable; follicles were no larger than 2.5 cm; and there were no fibroids, adenomyosis, endometrial thickening or fluid, or a subjectively increased amount of pelvic fluid (12,13).
The US technique used was transabdominal in 54 patients, endovaginal in 10 patients, and both in 22 patients. All studies were performed with a model 3000 (Advanced Technology Laboratories, Bothell, Wash) or XP-128 (Acuson, Mountain View, Calif). The US images were interpreted by one of five fellowship-trained radiologists (including R.D.H.). All images were retrospectively reviewed by a single sonologist (R.D.H.) to confirm the US inclusion criteria. The investigational review board approved the study. Verbal informed consent was obtained from the patients over the telephone prior to the interview.
We did not attempt to grade or quantify the pelvic pain on any objective or subjective scale but assumed the pain was clinically important enough to warrant referral for pelvic US. Patients with intermittent, continuous, unilateral or diffuse pelvic pain were included.
Additional data analyzed included the duration of pain before US, reports of any further imaging (computed tomography [CT] or US) after the initial US examination, and any subsequent surgical procedures (uterine dilation and curettage, endometrial biopsy or ablation, hysteroscopy, pelvic laparoscopy, or hysterectomy).
Patient follow-up data at 6–21 months (mean, 14 months) were available. Data were analyzed with STATA statistical software (Stata, College Station, Tex). Separate t test and 
2 analyses were conducted when appropriate for patients in the various outcome groups. For all analyses, a two-tailed P value less than .05 was considered to indicate a statistically significant difference. Regression analysis was performed for any predictors of outcome of pelvic pain.
The disease discovered at surgery was retrospectively graded by two radiologists (R.D.H., S.R.H.) in consensus as major, that is, capable of causing pain, or as minor, that is, unlikely to cause pain. Major disease included endometriosis, adenomyosis, and adhesions. Minor disease included an endometrial polyp or small fibroid.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The outcome of the pelvic pain was complete resolution in 34 of 86 (40%), improvement in 32 of 86 (37%), and persistence or worsening in 20 of 86 (23%) (Fig 1). The mean age of the patients in the groups with an outcome of improvement or resolution (mean age, 32 years) was not significantly different from that of the patients in the group with persistent pain (mean age, 30 years) (P = .68). The mean duration of pain prior to US for those female patients whose pain resolved or improved was 3.5 months. This was significantly different from the mean duration of pain for those female patients whose pain persisted (mean duration, 7.4 months; P = .02).
|
|
Six patients underwent laparoscopy: Three had normal findings, one had pelvic adhesions, one had "minimal endometriosis" (surgical report wording), and one had endometriosis and pelvic varices. Pain outcomes for this group were resolution in two patients, improvement in three patients, and persistence in one patient.
Six female patients underwent hysterectomy. Two patients had no disease identified. Two patients had "minor" (unlikely cause of pelvic pain) disease: One patient had a "small" (no size reported) endometrial polyp, and one had a 1-cm fibroid. Two patients had "major" disease: One patient had a 7-mm focus of adenomyosis (Fig 3), and one patient had endometriosis and pelvic varices. In this subset of patients, three had complete resolution of pain, and three had improvement.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The magnitude of this problem is immense. It is estimated that gynecologic surgery for pelvic pain performed in the United States costs more than $2 billion annually (14). The most frequent causes of pelvic pain are endometriosis, pelvic adhesions, and ovarian cysts (15). Some causes of pelvic pain may be straightforward to diagnose, such as a functional ovarian cyst. Much pelvic pain, however, has an organic basis that may be difficult to detect, as well as complex psychosocial components (14).
Most studies categorize pelvic pain as acute (duration < 3–6 months) or chronic (duration 
3–6 months) (15). Causes of acute pain are usually related to pregnancy, functional ovarian cysts, infections (pelvic inflammatory disease, including chlamydia), or abscesses (16–20). The first two entities and tubo-ovarian abscesses are frequently diagnosed with US; pelvic inflammatory disease is usually not diagnosed with US (2,3,15,18–20). The differential diagnosis for chronic pelvic pain includes adhesions, endometriosis, adenomyosis, endosalpingiosis, infections, and neoplasms (4,15,21,22). The first five diagnoses are often difficult to detect reliably with US, and the diagnosis of ovarian tumors, while improving with advances in endovaginal and color Doppler US techniques, may still be problematic because of the overlapping sonographic appearances of benign and malignant disease (9,23,24).
In our US section, pelvic pain accounts for approximately 25% of the referrals for pelvic US. The germinal idea of this study was found in the large number of normal findings of pelvic US examinations performed for this indication. In addition, with the avalanche of medical practice appropriateness criteria (largely due to managed care) it has become more common for third-party payers to investigate patient outcome data with regard to imaging studies.
The outcome of pelvic pain in patients with negative pelvic US findings in our series was predominantly favorable. Our results agree with those of the study by Barloon et al (11) with 59 patients referred for endovaginal US for a variety of clinical indications (approximately one-half for pelvic or right-lower-quadrant abdominal pain). These authors found the negative predictive value of normal endovaginal US findings to be 92% (54 of 59) with either (a) clinical follow-up (at least 1 year after the examination) or (b) surgery or endoscopy. We had a similar positive outcome result (negative surgical findings and resolved or improved clinical status) of 70%. Only one patient had demonstrable disease at subsequent imaging; this patient had acute diverticulitis, which may not have been present during the US examination 1 month before. Disease such as pelvic adhesions, small implants of endometriosis, and pelvic inflammatory disease are unlikely to be imaged with any present technique and require surgery and/or culture for diagnosis in most cases. Pelvic varices, tubo-ovarian abscesses, and adenomyosis may be seen with endovaginal US or magnetic resonance (MR) imaging (11,25–27)—although in our series, the focus of adenomyosis was small (7 mm) and was not detected at the transabdominal and endovaginal US examination performed in the patient. For this reason, we now perform endovaginal scanning after a negative transabdominal study in all patients with pelvic pain. The role of pelvic varices in pelvic pain has been proposed by Beard et al (28), but treatment is controversial and some authors even discount the diagnosis; this subject is beyond the scope of this article.
There seemed to be no dominant pattern to the referral for further imaging or surgery. Referral and selection bias were present, as this was a prospective, non–case-controlled study with a relatively small number of patients, and no attempt was made to account for referring physicians’ practices. A single academic medical center that also has a large community-based population was the source of all patients.
This study has several limitations. Few patients underwent additional imaging or surgery, so for the most part the outcomes were spontaneous or perhaps related to medical treatment, which we did not assess. We did not collect data on the positive pelvic US studies in the same period; therefore, we cannot calculate the sensitivity or the positive predictive value of an abnormal pelvic US scan. Also, we made no attempt to grade or categorize the patient’s pelvic pain because this is extremely complex and beyond the scope of our study. Rather, we attempted to investigate a common clinical scenario in a moderately busy (40–50-cases-per-day) US section and hope to provide radiologists (and gynecologists) with some pertinent information about what may happen to patients with a negative pelvic US study. Imaging, especially US, is an important part of the diagnostic work-up, but referral patterns and practices depend on the habits and experience of the referring physician. Some physicians have a low threshold for obtaining images, especially in acute situations, while others adopt a more conservative ("watchful waiting") approach and obtain images only as a last resort. In addition, pelvic US examinations were performed by one of three methods (transabdominal, endovaginal, or both). This variation was due to radiologist preference and patient features such as body habitus. The standard of care is to now perform endovaginal or both endovaginal and transabdominal examinations in all patients in whom this is possible. We also believe that the main outcome measures used are worthwhile but not optimal in that pain that has improved does not necessarily mean that a patient is satisfied with these results. To answer the important question of whether the patient is satisfied, future studies focused on quality of life are needed.
In conclusion, most female patients fulfilling the inclusion criteria of this study who had pelvic pain and a normal pelvic US scan will get better, with or without surgical intervention, and further imaging is unlikely to yield positive results. It appears that the majority of referring clinicians, at least in our institution, accept an US examination with negative findings as the final imaging test for most patients. A minority of patients who do undergo subsequent surgery will have clinically important disease, most of which US in its present technologic state is unlikely to depict.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. Kuligowska, L. Deeds III, and K. Lu III Pelvic Pain: Overlooked and Underdiagnosed Gynecologic Conditions RadioGraphics, January 1, 2005; 25(1): 3 - 20. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Cook, M. Quinn, and R. D. Harris Imaging in Women with Persistent Pelvic Pain Dr Harris responds: Radiology, May 1, 2001; 219(2): 574-a - 574. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
