- Email: [email protected]
Practice patterns among reproductive endocrinologists: further aspects of the infertility evaluation
FERTILITY AND STERILITYt VOL. 70, NO. 2, AUGUST 1998 Copyright ©1998 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A.
Practice patterns among reproductive endocrinologists: further aspects of the infertility evaluation Isaac Z. Glatstein, M.D.,*† Bernard L. Harlow, Ph.D.,‡ and Mark D. Hornstein, M.D.* Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts
Objective: To characterize infertility testing in clinical practice. Design: A cross-sectional study assessing differences in clinical practice based on identifiable practitioner variables, such as gender, age, and type of practice. Setting: A population-based national survey. Participant(s): U.S. board-certified reproductive endocrinologists. Main Outcome Measure(s): Physicians’ preferences in the interpretation and performance of the semen analysis, postcoital test (PCT), hysterosalpingography (HSG), laparoscopy, immunologic testing, hamster egg penetration assay, and screening for the luteinized unruptured follicle syndrome. Result(s): Younger physicians more often favor the use of Kruger strict morphology for the semen analysis, rely on urine or blood LH measurements for the timing of the PCT, and use a less strict cutoff for the definition of a normal PCT. About half of all physicians use antibiotic prophylaxis before HSG and 90% rely on water-soluble dye for imaging. Eighty percent of physicians in western U.S. practices schedule diagnostic laparoscopy during the follicular phase compared with 50% of their East Coast colleagues. Two thirds of physicians would conduct immunologic testing for infertility, but most would not use the hamster egg penetration test or screen for the luteinizing unruptured follicule syndrome.
Received October 22, 1997; accepted and revised February 10, 1998. Supported by a grant from Serono Laboratories, Inc., Randolph, Massachusetts. Reprint requests: Bernard L. Harlow, Ph.D., Obstetrics and Gynecology Epidemiology Center Brigham and Women’s Hospital, 221 Longwood Avenue, Boston, Massachusetts 02115 (FAX: 617-732-4899; E-mail: [email protected] .harvardb.edu). * Department of Obstetrics, Gynecology, and Reproductive Biology. † Present address: Deaconess Waltham Hospital, Waltham, Massachusetts. ‡ Obstetrics and Gynecology Epidemiology Center. 0015-0282/98/$19.00 PII S0015-0282(98)00134-4
Conclusion(s): Although most board-certified reproductive endocrinologists agree on the major areas of the performance of infertility testing, there was significant variability in the details of the performance of most testing, especially with respect to physician age and geographic location. (Fertil Sterilt 1998;70:263–9. ©1998 by American Society for Reproductive Medicine.) Key Words: Infertility, cross-sectional studies, surveys, physicians, reproductive endocrinology
Infertility patients in the United States are cared for by a wide spectrum of physicians, including family practitioners, internists, endocrinologists, obstetricians and gynecologists, urologists, and reproductive endocrinologists. Despite widely practiced treatment of infertility, it is agreed universally that the cornerstones of assessing an infertile couple are grouped into five standard testing categories. These include an assessment of ovulation, the semen analysis, an evaluation of uterine architecture and tubal patency, the postcoital test (PCT), and laparoscopy. We recently reported, in a nationwide survey of U.S. board-certified reproductive endocrinologists (an easily identifiable group of physicians with specialized training in infertility), that close to 80% of physicians routinely order each of these tests in their assessment of the infertile couple (1). We further reported
that, although most reproductive endocrinologists agree on the basics of the investigation, there was considerable variation in the use of other infertility testing based on identifiable physician characteristics, such as age, sex, practice type, and demographics. Our initial report focused on the basic testing that physicians routinely order in the course of an infertility investigation. The goal of the present study is to extend our previous work and to report the specifics of how infertility testing is performed and interpreted by reproductive endocrinologists.
MATERIALS AND METHODS Details pertaining to the survey of all U.S. board-certified reproductive endocrinologists are described in detail elsewhere (1). Briefly, self-administered interviews were completed 263
TABLE 1 Percent of board-certified reproductive endocrinologists who use certain procedures and guidelines in assessing the results of semen analyses during the standard infertility evaluation by physician sex and age, size and type of clinical practice, and geographic region. Semen evaluation manual count
Abstention $2 days Physician/practice characteristics All (n 5 398) Sex of physician Male (n 5 334) Female (n 5 64) Age (y) $50 (n 5 99) 40–49 (n 5 204) ,40 (n 5 95) Patients per month ,20 (n 5 211) $20 (n 5 187) Practice setting University (n 5 217) Private (n 5 181) Geography Northeast (n 5 133) Midwest (n 5 186) West (n 5 78)
Percent
OR* (95% CI)
87.4
Percent
OR* (95% CI)
60.3
Sperm concentration $20 3 106
Percent
OR* (95% CI)
91.6
Sperm motility $50%
Percent
OR* (95% CI)
66.8
Type of morphology basic
Percent
OR* (95% CI)
66.4
86.8 89.1
1.0 — 1.3 (0.5–3.1)
60.2 60.9
1.0 — 1.1 (0.6–1.9)
90.7 92.2
1.0 — 1.3 (0.4–4.6)
66.5 68.8
1.0 — 1.3 (0.7–2.3)
62.9 64.1
1.0 — 1.2 (0.7–2.3)
81.8 84.8 94.5
1.0 — 1.1 (0.6–2.2) 4.5 (1.4–14.2)
56.6 62.8 59.0
1.0 — 1.3 (0.8–2.2) 1.2 (0.7–2.2)
78.8 93.6 97.9
1.0 — 3.9 (1.7–8.7) 10.9 (2.4–50.0)
72.7 66.2 62.1
1.0 — 0.7 (0.4–1.3) 0.7 (0.3–1.2)
74.7 59.8 57.9
1.0 — 0.5 (0.3–0.9) 0.4 (0.2–0.8)
83.4 89.8
1.0 — 1.7 (0.9–3.3)
60.7 59.9
1.0 — 0.9 (0.6–1.3)
89.1 93.1
1.0 — 1.0 (0.5–2.3)
68.2 65.2
1.0 — 0.8 (0.5–1.3)
67.3 58.3
1.0 — 0.7 (0.4–1.1)
82.0 91.7
1.0 — 2.7 (1.4–5.4)
55.3 66.3
1.0 — 1.6 (1.1–2.5)
91.2 90.6
1.0 — 1.1 (0.5–2.5)
65.9 68.0
1.0 — 1.2 (0.8–1.8)
65.4 60.2
1.0 — 0.9 (0.6–1.5)
88.7 85.0 87.2
1.0 — 0.8 (0.4–1.6) 0.9 (0.4–2.2)
53.7 64.5 61.5
1.0 — 1.5 (1.0–2.5) 1.2 (0.7–2.2)
94.8 90.9 84.6
1.0 — 0.3 (0.1–1.1) 0.2 (0.05–0.6)
62.7 74.2 56.4
1.0 — 1.7 (1.0–2.8) 0.7 (0.4–1.3)
58.2 66.1 64.1
1.0 — 1.3 (0.8–2.2) 1.2 (0.7–2.3)
p Odds ratios adjusted for all other factors in the table.
by 398 (84.1%) of 473 eligible physicians. The survey contained three distinct sections. The first section queried the background of each physician, including age, sex, type of training, affiliation, size, and geographic location of the infertility practice. We previously reported data from the second section, which exclusively dealt with basics of the infertility workup including which tests are ordered during the course of an evaluation of an infertile couple (1). The third section, which is the subject of this report, involves the specific details of the performance of each of the main infertility diagnostic testing categories. This section includes the semen analysis, the PCT, hysterosalpingography (HSG), laparoscopy, and miscellaneous tests and practices. As in our previous analysis, we determined rates of diagnostic test use and determined whether these rates varied by the age or sex of the clinician, the size or academic affiliation of the practice, and the geographic location of the practice office. Differences in rates of use were assessed through the use of Cochran-Mantel-Haenszel x2 analyses after adjustment for specific physician background characteristics.
RESULTS Semen Analysis Questions pertaining to semen analyses include reproductive endocrinologists’ recommendations concerning length 264
Glatstein et al.
Infertility evaluation practice patterns
of abstinence before producing the sample for semen analysis, the method used to evaluate a semen sample (i.e., computer-assisted versus a manual count), and the clinicianpreferred cutoff values for a normal sperm concentration and motility. In addition, we queried whether physicians preferred to rely on the Kruger (strict) or the basic morphological evaluation. Data generated from these questions are shown in Table 1. Overall, approximately 87% of reproductive endocrinologists preferred an abstention period of $2 days before providing a semen analysis. The remaining endocrinologists indicated either no recommendation or an abstention period of #1 day. The 2-day abstention period was preferred more strongly among physicians ,40 compared with those $40 years of age as well as among those in private practice compared with university-based practices. Sixty percent of reproductive endocrinologists used a manual count method rather than a computer-assisted method for the evaluation of the semen sample. Overall, 92% of clinicians considered the cutoff for a normal sperm count to be at least 20 3 106 sperm/mL. There was less uniformity among physicians over age 50 with .20% preferring semen cutoffs other than 20 3 106, ranging from as low as 10 3 106 to as high as 60 3 106. Virtually all (98%) physicians ,40 years of age preferred the 20 3 106 cutoff for sperm concentration. Vol. 70, No. 2, August 1998
Compared with clinicians from the Northeast, physicians from the Midwest varied more in their selected sperm concentration cutoffs, with the greatest variation seen among clinicians from the West Coast. Approximately 67% of reproductive endocrinologists considered 50% the minimum motility cutoff, whereas 28% were satisfied with at least 40% motility. Only 5% of clinicians considered 30% the desired cutoff for normal motility. Although there was wide variation in normal sperm motility standards, there was little variation by characteristics of the physician or their practice. Approximately 66% of reproductive endocrinologists stated that they rely on the basic morphology, whereas the remainder chose the Kruger system. There did, however, appear to be a weak trend for younger physicians to lean toward use of the Kruger morphology. The cutoff values for normal morphology preferred by clinicians in western U.S. practices tended to be higher, regardless of their preference for basic or Kruger assessment. For example, 70% of western clinicians preferring the basic morphology relied on a normal morphology cutoff of $50% compared with 54% of eastern clinicians. Likewise, 82% of western clinicians preferring the Kruger morphology relied on a normal morphology cutoff of $14% compared with 62% of eastern clinicians.
Postcoital Test We assessed details pertaining to physician practices concerning the PCT. Areas explored included preferred type of PCT (standard Sims-Huhner versus fractional), the length of abstinence before its performance, the determination of timing, whether the posterior vaginal fornix is aspirated for sperm, and the minimum motile sperm number for a normal PCT. Analysis of the PCT data revealed that 81.4% of all respondents normally order this test in the evaluation of the infertile couple. There was little variation in this trend by sex or geographic location of the clinician. Ninety percent of clinicians $50 years of age regularly ordered PCTs compared with 72% of younger physicians (P 5 .03), and 92% of private practitioners regularly ordered PCTs compared with 73% of their university-affiliated associates (P,.001). These significant differences remained after multivariate adjustment for all clinician characteristics. Among the selected population of clinicians who regularly ordered PCT, 89.7% reported that they relied on the standard Sims-Huhner test rather than the fractional PCT. Table 2 displays the data concerning the minimum motile sperm cutoff for a normal PCT and the methodology used to determine the optimum timing of the PCT. There was a significant trend of clinicians $50 years of age to select a greater minimum cutoff for a normal PCT in terms of sperm visualized per high-power field, whereas clinicians ,50 years of age gravitated toward allowing a lesser number of sperm present on the PCT to qualify it as a “normal” test (P 5 0.06). Regarding the test used to determine the optimal timing of the PCT, older physicians clearly relied more on cycle length estimation, whereas younger physicians relied FERTILITY & STERILITYt
TABLE 2 Percent of board-certified reproductive endocrinologists who order the PCT and use various methods for determining the timing of the PCT during the standard infertility evaluation by physician sex and age. Age of clinician (y)
Postcoital test characteristics Minimum cutoff for a normal PCT (sperm/HPF) $1 $3 $5 Cochran-Mantel-Haenszel x 2* Tests used to determine the timing of PCT Urine/blood LH Basal body temperature Cycle lengths Ultrasound to assess follicular growth Cochran-Mantel-Haenszel x 2*
$50
40–49
,40
No. (%)
No. (%)
No. (%)
15 (17.7) 42 (49.4) 28 (32.9)
26 (31.7) 14 (17.1) 35 (42.7) 7 (8.5)
55 (35.7) 25 (32.9) 78 (50.7) 40 (52.6) 21 (13.6) 11 (14.5) 14.4, P 5 .006
78 (52.4) 8 (5.4) 43 (28.9)
40 (53.3) 4 (5.3) 21 (28.0)
20 (13.4) 10 (13.3) 15.7, P 5 0.15
Note: HPF 5 high-power field. p Adjusted for type of practice (private, university), size of practice (,20, $20 patients), and geographic location (East, Midwest, West).
more often on urine or blood LH determinations (P 5 .015). Of clinicians who regularly ordered PCT assessment, 77% reported that they regularly aspirated the posterior vaginal fornix before the performance of the PCT. This distribution did not vary by clinician characteristics or practice location.
Hysterosalpingography The results of practice patterns involving the performance of the HSG are shown in Table 3. There was virtually no variation in the interpretation of the results by age or sex of the clinician. We observed that private practitioners more frequently performed HSG studies on their own without the assistance of a gynecologist or radiologist compared with university-affiliated colleagues. In addition, there was a significant geographic variability, with a greater proportion of physicians on the East Coast primarily performing their own HSG compared with practitioners in the West who tended to rely more heavily on their radiologist colleagues. Most practitioners on the East Coast and in the Midwest tended to use water-soluble media. However, although most practitioners in the western United States used water-soluble media, there was a significantly greater proportion of clinicians performing a water-soluble HSG followed by an oilbased dye “chaser” than that observed among physicians from both the Midwest and the East Coast. It is of interest that 70% of practitioners did not provide their patients with antibiotic prophylaxis before the performance of the HSG. However, physicians in private practice and those in the 265
TABLE 3 Percent of board-certified reproductive endocrinologists who order HSG and their methods for interpreting the results as part of the standard infertility evaluation by type of clinical practice and geographic region. Practice setting
HSG characteristics Clinician who generally performs HSG Self Colleague Radiologist Cochran-Mantel-Haenszel x 2* Contrast media used in performing HSG Oil-based media Water-soluble media Water-soluble followed by oil-based Cochran-Mantel-Haenszel x 2* Antibiotic prophylaxis in patients before HSG No Yes Cochran-Mantel-Haenszel x 2*
Practice location
University
Private
East
Midwest
West
No. (%)
No. (%)
No. (%)
No. (%)
No. (%)
123 (57.2) 138 (77.5) 43 (20.0) 4 (2.3) 49 (22.8) 36 (20.2) 28.3, P,.001
85 (64.4) 24 (18.2) 23 (17.4)
138 (75.0) 20 (10.9) 26 (14.1) 46.1, P,.001
38 (49.4) 3 (3.9) 36 (46.8)
14 (6.5) 15 (8.5) 183 (85.1) 146 (82.5) 18 (8.4) 16 (9.0) 0.33, P 5 .85
7 (5.3) 118 (89.4) 7 (5.3)
15 (8.2) 159 (86.4) 10 (5.4) 19.3, P 5 .001
7 (9.2) 52 (68.4) 17 (22.4)
160 (74.4) 114 (64.0) 55 (25.6) 64 (36.0) 3.1, P 5 .08
102 (77.3) 30 (22.7)
131 (71.2) 53 (28.8) 6.6, P 5 .04
41 (53.3) 36 (46.8)
p Adjusted for physician sex, age, and size of practice (,20, $20 patients).
western portion of the United States did tend to provide antibiotic coverage more often. Of those physicians who did prescribe prophylaxis before the HSG, virtually all chose a tetracycline-based regimen as their preferential choice. Only three clinicians chose a cephalosporin-based regimen, and two chose metronidazole-based regimens (data not shown).
Laparoscopy Overall, 64% of physicians chose to schedule laparoscopy during the follicular phase of the cycle (Table 4). This rate varied substantially by geographic region, with .80% of clinicians from the western United States preferring follicular phase laparoscopy and only half of the eastern physicians expressing the same preference. Most clinicians (67%) chose not to conduct hysteroscopy at the time of laparoscopy. However, all but 14 clinicians performed tubal lavage at the time of laparoscopy. Relative to university-based practitioners, there was a trend toward physicians in private practice to add a hysteroscopic evaluation in addition to laparoscopic assessment (P 5 .02). This trend pertained to a third of clinicians in private practice with only 17% of university-based clinicians offering the same prophylaxis. Of those clinicians who did provide regular antibiotic prophylaxis, 70% selected cephalosporin as their first choice with the remaining preferring a tetracycline-based regimen.
Immunologic Infertility Testing The final section of this survey inquired about miscellaneous tests and practices, including the assessment of immunologic infertility, the conditions under which a hamster egg sperm penetration assay is used, and the routine screen266
Glatstein et al.
Infertility evaluation practice patterns
ing of infertility patients for the luteinized, unruptured follicle syndrome. Approximately 72% of respondents stated they would conduct immunologic infertility testing under the appropriate circumstances with a somewhat greater proportion of private practitioners and western U.S. practitioners favoring this type of testing (Table 5). Of those practitioners who conduct immunologic testing, 95% stated that they would rely primarily on the immunobead method. Seven respondents said they would rely on ELISA testing, five on sperm agglutination, and two on a complement-dependent assay. Wide variation existed among clinicians who would or would not carry out hamster egg sperm penetration assays by practice setting and geographic location. With respect to using these assays for assessing abnormal semen analyses, 30% of private reproductive endocrinologists felt it was appropriate compared with 15% of university-based clinicians. Little difference was observed between private and university-based reproductive clinicians in the proportion of those who believed that these assays were appropriate for the assessment of unexplained infertility and as an IVF prognostic guide. In general, a significantly greater proportion of western U.S. clinicians believed that hamster egg sperm penetration assays was appropriate to use for assessing abnormal semen analyses, unexplained infertility, and as a prognostic IVF guide compared with both eastern and midwestern clinicians. As observed for the hamster egg sperm penetration assays, a significantly greater proportion of private practitioners (23.5%) reported that they routinely screen patients for Vol. 70, No. 2, August 1998
TABLE 4 Procedures followed by board-certified reproductive endocrinologists regarding laparoscopy as part of the standard infertility evaluation by type of clinical practice and geographic region. Practice setting
Laparoscopy characteristics When laparoscopy usually scheduled Follicular phase Luteal phase Scheduling availability Cochran-Mantel-Haenszel x 2* Hysteroscopy done at time of laparoscopy No Yes Cochran-Mantel-Haenszel x 2p Antibiotic prophylaxis in patients before laparoscopy No Yes Cochran-Mantel-Haenszel x 2*
Practice location
University
Private
East
Midwest
West
No. (%)
No. (%)
No. (%)
No. (%)
No. (%)
133 (61.6) 121 (67.6) 11 (5.1) 17 (9.5) 72 (33.3) 41 (22.9) 7.2, P 5 .03
70 (52.6) 10 (7.5) 53 (39.9)
122 (66.1) 15 (8.1) 48 (25.9) 19.3, P 5 .001
62 (80.5) 3 (3.9) 12 (15.6)
158 (72.8) 109 (60.6) 59 (27.2) 71 (39.4) 5.1, P 5 .02
87 (65.4) 46 (34.6)
127 (68.3) 59 (31.7) 0.29, P 5 .87
53 (68.0) 25 (32.0)
179 (82.5) 119 (66.1) 38 (17.5) 61 (33.9) 13.9, P,.001
106 (79.7) 27 (20.3)
136 (73.1) 50 (26.9) 3.6, P 5 .17
56 (71.8) 22 (28.2)
p Adjusted for physician sex, age, and size of practice (,20, $20 patients).
luteinized unruptured follicles compared with only 8% of university-based clinicians. Although western U.S. practitioners did screen for this syndrome more often than eastern or midwestern practitioners, the difference was substantially less than that observed for hamster egg penetration assays.
DISCUSSION Our first area of questioning concerned semen analysis. It appears that physicians in a private practice setting have a significantly higher rate of recommending to patients that the semen analysis be produced after $2 days of abstention. This was not seen in all the physicians and those in a university practice. It is of interest that the recommendations put forth by the World Health Organization (WHO) state that “ideally the sample should be collected after a minimum of 48 hours but not longer than seven days of sexual abstinence” (2). It is unclear why younger physicians and those in private practice would be more strictly adherent to these guidelines, but this may be a result of fellowship training practices, which only recently have been introduced.
morphological examination, i.e., the basic (WHO) versus the Kruger strict determination of morphology. Although 66% of all respondents stated that they did favor the basic morphology, there was a statistically significant trend among younger physicians to favor the Kruger morphology. Of interest is a recent study by Morgentaler et al. (3) that directly compared the WHO and strict criteria methodologies using IVF outcome as an endpoint. In this study the traditional basic morphology demonstrated a higher sensitivity and negative predictive value compared with strict criteria. We would suggest that the modern evaluation of sperm is an evolving process, and given the rapid developments of microfertilization and intracytoplasmic sperm injection (ICSI), need to be reexamined and reevaluated in light of these new technologies.
Overall, approximately 60% of physicians prefer the manual count when performing an evaluation of a semen sample compared with a computer-assisted analysis. There was, however, no discernible difference between the various identifiable individual physician groups. The use of the computer-assisted semen analysis is an evolving field, and as the methods are perfected and equipment becomes more affordable and readily available, we may expect to see practice patterns shift in this regard.
The PCT is an area of infertility testing that has engendered much controversy in recent years. For example, perhaps one of the most vexing questions surrounding its use is the cutoff that defines a normal test. This number has varied in studies from a high of 20 sperm per high-power field to as low as one moving spermatozoa (4). Our data demonstrate that although older physicians tend to have a higher threshold for determining a satisfactory PCT, younger physicians tend to allow fewer numbers of sperm to be present to qualify it as a normal test. Likewise, younger physicians tended to rely on more modern methods of ovulation detection, such as urine or blood LH assessment to determine the timing of the PCT, versus older physicians who tended to rely more heavily on traditional methods of estimating ovulation such as cycle lengths.
Another unsettled topic of semen analysis is the type of
These differences may reflect simply the phenomenon of
FERTILITY & STERILITYt
267
TABLE 5 Percent of board-certified reproductive endocrinologists who test for immunologic infertility, conditions under which practitioners use hamster egg sperm penetration, and whether practitioners screen infertility patients for luteinized unruptured follicle syndrome by type of clinical practice and geographic region. Practice setting
Other infertility tests and procedures Immunologic infertility testing Would not test Would test Cochran-Mantel-Haenszel x 2* Conditions under which hamster egg sperm penetration assays are used Abnormal semen tests Would not test Would test Cochran-Mantel-Haenszel x 2* Unexplained infertility Would not test Would test Cochran-Mantel-Haenszel x 2* Prognostic IVF guide Would not test Would test Cochran-Mantel-Haenszel x 2* Clinicians that routinely screen for luteinized unruptured follicle syndrome Would not screen Would screen Cochran-Mantel-Haenszel x 2*
Practice location
University
Private
East
Midwest
West
No. (%)
No. (%)
No. (%)
No. (%)
No. (%)
73 (33.6) 39 (21.6) 144 (66.4) 142 (78.4) 4.5, P 5 .04
48 (36.1) 85 (63.9)
48 (25.8) 138 (74.2) 8.0, P 5 .02
16 (20.5) 62 (79.5)
182 (84.7) 127 (70.2) 33 (15.4) 54 (29.8) 12.2, P,.001
113 (85.0) 20 (15.0)
151 (82.1) 33 (17.9) 23.8, P,.001
45 (57.7) 33 (42.3)
141 (65.6) 105 (58.0) 74 (34.4) 76 (42.0) 2.9, P 5 .08
98 (73.7) 35 (26.3)
122 (66.3) 62 (33.7) 35.0, P,.001
26 (33.3) 52 (66.7)
153 (71.2) 123 (68.0) 62 (28.8) 58 (32.0) 0.28, P 5 .6
111 (83.5) 22 (16.5)
122 (66.3) 62 (33.7) 20.8, P,.001
42 (53.9) 36 (46.1)
199 (92.1) 137 (76.5) 17 (7.9) 42 (23.5) 18.0, P,.001
122 (92.4) 10 (7.6)
149 (81.0) 35 (19.0) 5.8, P 5 .06
64 (82.1) 14 (18.0)
* Adjusted for physician sex, age, and size of practice (,20, $20 patients).
younger, more recently trained physicians, using more recently introduced technology. This is a concept that has been well-documented in many areas of medicine including data that we previously reported (1). The subject of oil- versus water-soluble contrast media in the performance of the hysterosalpingogram has long been a matter of debate in the reproductive endocrinology literature. There are proponents of both approaches, with those who favor oil contrast quoting literature demonstrating that oilsoluble contrast media has a therapeutic effect in regards to pregnancy rates compared with water-soluble media. A metaanalysis by Watson and coworkers (5) encompassing four randomized clinical trials and six nonrandomized controlled studies evaluated pregnancy rates after the use of each of these contrast media. They concluded that the odds ratio for pregnancy was 1.9 for the oil-based dye compared with the use of water-soluble contrast media. In addition to this important increase in the probability of pregnancy with oil media, proponents of the oil method also claim that the incidence of abdominal pain, vaginal bleeding, and pelvic inflammatory disease (PID) is significantly lower compared with that observed in women experiencing watersoluble media. Loy et al. (6) found that the sensitivity and predictive value of HSG was higher in oil-soluble media 268
Glatstein et al.
Infertility evaluation practice patterns
compared with water-soluble media when using a laparoscopic diagnosis as the gold standard. Opponents of oilbased contrast media cite the danger of a potential increased risk of embolism if intravasation occurs, and the concern that if oil is trapped within a hydrosalpinx, a granuloma may be formed. However, with modern fluoroscopic methods of performing the HSG, the risk of embolism virtually is eliminated. In addition, the issue of granuloma formation of oil media within tubes is somewhat theoretical. With this in mind, it is somewhat difficult to explain why most physicians choose to continue to use water-soluble media, given the relative advantages of the oil method. At the very minimum, a water-soluble hysterosalpingogram, followed by an oil “chaser” if the tubes are patent, would provide the advantages of oil without the disadvantages of potential trapping of oily fluid within a hydrosalpinx. It may simply be that the metaanalysis study, which came out of the United Kingdom, has not been disseminated widely within training programs and the reproductive endocrine community. Concerning antibiotic prophylaxis, our data show that most physicians do not provide antibiotic prophylaxis routinely for HSG. Although the literature concerning the use of antibiotic prophylaxis and HSG is somewhat scarce, it would Vol. 70, No. 2, August 1998
seem logical that, given the devastating consequences of PID after a hysterosalpingogram, it is certainly not unreasonable to recommend antibiotic prophylaxis, given the relatively low morbidity associated with a short course of prophylaxis. A recently published bulletin by the American College of Obstetricians and Gynecologists likewise recommends antibiotic coverage in all patients undergoing an HSG procedure (7). Laparoscopy generally is considered a basic element of the infertility workup, although its routine use in cases in which there is no compelling evidence for pelvic disease and in which the hysterosalpingogram is normal, may not be cost-effective. Most physicians schedule the laparoscopy during the follicular phase of the cycle. This is logical because there is less of a chance of inadvertently performing a laparoscopy during pregnancy. Although most physicians who were surveyed do not perform a hysteroscopy routinely at the time of laparoscopy, there was a trend for physicians in private practice to offer this procedure compared with their university-based colleagues. Immunologic testing for infertility is a controversial topic. It is of interest that physicians on the West Coast tended to order more immunologic testing than their East Coast colleagues. This may be reflective of fellowship and infertility training practices on the West Coast. Except for physicians residing on the West Coast, when encountering patients with unexplained infertility, most physicians on the East Coast and in the Midwest stated that they would not routinely order hamster egg sperm penetration assay. A study using life-table analyses and monthly fecundity rates by O’Shea et al. (8) has shown that the hamster egg penetration score is not predictive of the incidence of pregnancy nor of time to conception. It would appear that, especially with the more prevalent use of IVF and ICSI for male factor, the hamster egg sperm penetration assay generally has fallen out of favor with reproductive endocrinologists. The final category of infertility testing concerns routine screening for the luteinized, unruptured follicle syndrome. Most clinicians stated that they would not screen for this condition. We may speculate that the reason for this lack of enthusiasm is the controversy over whether this entity actually exists and how one tests for it. Is the diagnosis made with laparoscopy with visualization of ovarian stigma, progesterone determinations, or the lack of follicular reduction
FERTILITY & STERILITYt
in size on ultrasound? (9, 10). Until this condition is defined more clearly and rigid criteria for its diagnosis are determined, it will remain most likely a marginal and elusive entity at best. In summary, we found that there were distinct practice pattern differences among infertility practitioners, especially with regards to age and geographic location. Younger, board-certified, more recently trained physicians tended to incorporate more modern methodologies into their infertility practice. Data such as these regarding the actual infertility workup as practiced by board-certified specialists is an important first step in seeking to better use resources in the treatment of infertile couples. These results may assist those who may seek to develop national standards in the evaluation of infertility. It will be interesting to observe how these practice patterns will change over time as new technologies become integrated into the procedures available to infertility practitioners.
Acknowledgment: The authors thank Ms. Paula Ouellette for assistance in the preparation of this manuscript.
References 1. Glatstein IZ, Harlow BL, Hornstein MD. Practice patterns among reproductive endocrinologists: the infertility evaluation. Fertil Steril 1997;67:443–51. 2. World Health Organization. Laboratory manual for the examination of human semen and semen-cervical mucus interaction. 3rd ed. New York: Cambridge University Press, 1992:3. 3. Morgentaler A, Fung MY, Harris DH, Powers RD, Alper MM. Sperm morphology and in-vitro fertilization outcome: a direct comparison of World Health Organization and strict criteria methodologies. Fertil Steril 1995;64:1177– 82. 4. Griffith CS, Grimes DA. The validity of the post-coital test. Am J Obstet Gynecol 1990;162:615–20. 5. Watson A, Vandekerchhove P, Lilford R, Vail A, Brosens I, Hughes E. A meta-analysis of the therapeutic role of oil-soluble contrast media at hysterosalpingography: a surprising result? Fertil Steril 1994;61:470 –7. 6. Loy RA, Weinstein FG, Seibel MM. Hysterosalpingography in perspective: the predictive value of oil-soluble versus water-soluble contrast media. Fertil Steril 1989;51:170 –2. 7. The American College of Obstetricians and Gynecologists. Antibiotics and gynecologic infections. ACOG Educational Bulletin No 237, June 1997. 8. O’Shea DL, Odem RR, Cholewa C, Gast MJ. Long term follow-up of couples after hamster egg penetration testing. Fertil Steril 1993;60: 1040 –5. 9. Hamilton MPR, Fleming R, Coutts JRT, Macaughton MC, Whitfield CR. Luteal cysts and unexplained infertility: biochemical and ultrasonic evaluation. Fertil Steril 1990;50:32–7. 10. Scheenges E, Velde ER, Kremer J. Inspection of the ovaries and steroids in serum and peritoneal fluid at various time intervals after ovulation in fertile women: implications for the luteinized unruptured follicle syndrome. Fertil Steril 1990;54:38 – 41.
269
Practice patterns among reproductive endocrinologists: further aspects of the infertility evaluation Isaac Z. Glatstein, M.D.,*† Bernard L. Harlow, Ph.D.,‡ and Mark D. Hornstein, M.D.* Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts
Objective: To characterize infertility testing in clinical practice. Design: A cross-sectional study assessing differences in clinical practice based on identifiable practitioner variables, such as gender, age, and type of practice. Setting: A population-based national survey. Participant(s): U.S. board-certified reproductive endocrinologists. Main Outcome Measure(s): Physicians’ preferences in the interpretation and performance of the semen analysis, postcoital test (PCT), hysterosalpingography (HSG), laparoscopy, immunologic testing, hamster egg penetration assay, and screening for the luteinized unruptured follicle syndrome. Result(s): Younger physicians more often favor the use of Kruger strict morphology for the semen analysis, rely on urine or blood LH measurements for the timing of the PCT, and use a less strict cutoff for the definition of a normal PCT. About half of all physicians use antibiotic prophylaxis before HSG and 90% rely on water-soluble dye for imaging. Eighty percent of physicians in western U.S. practices schedule diagnostic laparoscopy during the follicular phase compared with 50% of their East Coast colleagues. Two thirds of physicians would conduct immunologic testing for infertility, but most would not use the hamster egg penetration test or screen for the luteinizing unruptured follicule syndrome.
Received October 22, 1997; accepted and revised February 10, 1998. Supported by a grant from Serono Laboratories, Inc., Randolph, Massachusetts. Reprint requests: Bernard L. Harlow, Ph.D., Obstetrics and Gynecology Epidemiology Center Brigham and Women’s Hospital, 221 Longwood Avenue, Boston, Massachusetts 02115 (FAX: 617-732-4899; E-mail: [email protected] .harvardb.edu). * Department of Obstetrics, Gynecology, and Reproductive Biology. † Present address: Deaconess Waltham Hospital, Waltham, Massachusetts. ‡ Obstetrics and Gynecology Epidemiology Center. 0015-0282/98/$19.00 PII S0015-0282(98)00134-4
Conclusion(s): Although most board-certified reproductive endocrinologists agree on the major areas of the performance of infertility testing, there was significant variability in the details of the performance of most testing, especially with respect to physician age and geographic location. (Fertil Sterilt 1998;70:263–9. ©1998 by American Society for Reproductive Medicine.) Key Words: Infertility, cross-sectional studies, surveys, physicians, reproductive endocrinology
Infertility patients in the United States are cared for by a wide spectrum of physicians, including family practitioners, internists, endocrinologists, obstetricians and gynecologists, urologists, and reproductive endocrinologists. Despite widely practiced treatment of infertility, it is agreed universally that the cornerstones of assessing an infertile couple are grouped into five standard testing categories. These include an assessment of ovulation, the semen analysis, an evaluation of uterine architecture and tubal patency, the postcoital test (PCT), and laparoscopy. We recently reported, in a nationwide survey of U.S. board-certified reproductive endocrinologists (an easily identifiable group of physicians with specialized training in infertility), that close to 80% of physicians routinely order each of these tests in their assessment of the infertile couple (1). We further reported
that, although most reproductive endocrinologists agree on the basics of the investigation, there was considerable variation in the use of other infertility testing based on identifiable physician characteristics, such as age, sex, practice type, and demographics. Our initial report focused on the basic testing that physicians routinely order in the course of an infertility investigation. The goal of the present study is to extend our previous work and to report the specifics of how infertility testing is performed and interpreted by reproductive endocrinologists.
MATERIALS AND METHODS Details pertaining to the survey of all U.S. board-certified reproductive endocrinologists are described in detail elsewhere (1). Briefly, self-administered interviews were completed 263
TABLE 1 Percent of board-certified reproductive endocrinologists who use certain procedures and guidelines in assessing the results of semen analyses during the standard infertility evaluation by physician sex and age, size and type of clinical practice, and geographic region. Semen evaluation manual count
Abstention $2 days Physician/practice characteristics All (n 5 398) Sex of physician Male (n 5 334) Female (n 5 64) Age (y) $50 (n 5 99) 40–49 (n 5 204) ,40 (n 5 95) Patients per month ,20 (n 5 211) $20 (n 5 187) Practice setting University (n 5 217) Private (n 5 181) Geography Northeast (n 5 133) Midwest (n 5 186) West (n 5 78)
Percent
OR* (95% CI)
87.4
Percent
OR* (95% CI)
60.3
Sperm concentration $20 3 106
Percent
OR* (95% CI)
91.6
Sperm motility $50%
Percent
OR* (95% CI)
66.8
Type of morphology basic
Percent
OR* (95% CI)
66.4
86.8 89.1
1.0 — 1.3 (0.5–3.1)
60.2 60.9
1.0 — 1.1 (0.6–1.9)
90.7 92.2
1.0 — 1.3 (0.4–4.6)
66.5 68.8
1.0 — 1.3 (0.7–2.3)
62.9 64.1
1.0 — 1.2 (0.7–2.3)
81.8 84.8 94.5
1.0 — 1.1 (0.6–2.2) 4.5 (1.4–14.2)
56.6 62.8 59.0
1.0 — 1.3 (0.8–2.2) 1.2 (0.7–2.2)
78.8 93.6 97.9
1.0 — 3.9 (1.7–8.7) 10.9 (2.4–50.0)
72.7 66.2 62.1
1.0 — 0.7 (0.4–1.3) 0.7 (0.3–1.2)
74.7 59.8 57.9
1.0 — 0.5 (0.3–0.9) 0.4 (0.2–0.8)
83.4 89.8
1.0 — 1.7 (0.9–3.3)
60.7 59.9
1.0 — 0.9 (0.6–1.3)
89.1 93.1
1.0 — 1.0 (0.5–2.3)
68.2 65.2
1.0 — 0.8 (0.5–1.3)
67.3 58.3
1.0 — 0.7 (0.4–1.1)
82.0 91.7
1.0 — 2.7 (1.4–5.4)
55.3 66.3
1.0 — 1.6 (1.1–2.5)
91.2 90.6
1.0 — 1.1 (0.5–2.5)
65.9 68.0
1.0 — 1.2 (0.8–1.8)
65.4 60.2
1.0 — 0.9 (0.6–1.5)
88.7 85.0 87.2
1.0 — 0.8 (0.4–1.6) 0.9 (0.4–2.2)
53.7 64.5 61.5
1.0 — 1.5 (1.0–2.5) 1.2 (0.7–2.2)
94.8 90.9 84.6
1.0 — 0.3 (0.1–1.1) 0.2 (0.05–0.6)
62.7 74.2 56.4
1.0 — 1.7 (1.0–2.8) 0.7 (0.4–1.3)
58.2 66.1 64.1
1.0 — 1.3 (0.8–2.2) 1.2 (0.7–2.3)
p Odds ratios adjusted for all other factors in the table.
by 398 (84.1%) of 473 eligible physicians. The survey contained three distinct sections. The first section queried the background of each physician, including age, sex, type of training, affiliation, size, and geographic location of the infertility practice. We previously reported data from the second section, which exclusively dealt with basics of the infertility workup including which tests are ordered during the course of an evaluation of an infertile couple (1). The third section, which is the subject of this report, involves the specific details of the performance of each of the main infertility diagnostic testing categories. This section includes the semen analysis, the PCT, hysterosalpingography (HSG), laparoscopy, and miscellaneous tests and practices. As in our previous analysis, we determined rates of diagnostic test use and determined whether these rates varied by the age or sex of the clinician, the size or academic affiliation of the practice, and the geographic location of the practice office. Differences in rates of use were assessed through the use of Cochran-Mantel-Haenszel x2 analyses after adjustment for specific physician background characteristics.
RESULTS Semen Analysis Questions pertaining to semen analyses include reproductive endocrinologists’ recommendations concerning length 264
Glatstein et al.
Infertility evaluation practice patterns
of abstinence before producing the sample for semen analysis, the method used to evaluate a semen sample (i.e., computer-assisted versus a manual count), and the clinicianpreferred cutoff values for a normal sperm concentration and motility. In addition, we queried whether physicians preferred to rely on the Kruger (strict) or the basic morphological evaluation. Data generated from these questions are shown in Table 1. Overall, approximately 87% of reproductive endocrinologists preferred an abstention period of $2 days before providing a semen analysis. The remaining endocrinologists indicated either no recommendation or an abstention period of #1 day. The 2-day abstention period was preferred more strongly among physicians ,40 compared with those $40 years of age as well as among those in private practice compared with university-based practices. Sixty percent of reproductive endocrinologists used a manual count method rather than a computer-assisted method for the evaluation of the semen sample. Overall, 92% of clinicians considered the cutoff for a normal sperm count to be at least 20 3 106 sperm/mL. There was less uniformity among physicians over age 50 with .20% preferring semen cutoffs other than 20 3 106, ranging from as low as 10 3 106 to as high as 60 3 106. Virtually all (98%) physicians ,40 years of age preferred the 20 3 106 cutoff for sperm concentration. Vol. 70, No. 2, August 1998
Compared with clinicians from the Northeast, physicians from the Midwest varied more in their selected sperm concentration cutoffs, with the greatest variation seen among clinicians from the West Coast. Approximately 67% of reproductive endocrinologists considered 50% the minimum motility cutoff, whereas 28% were satisfied with at least 40% motility. Only 5% of clinicians considered 30% the desired cutoff for normal motility. Although there was wide variation in normal sperm motility standards, there was little variation by characteristics of the physician or their practice. Approximately 66% of reproductive endocrinologists stated that they rely on the basic morphology, whereas the remainder chose the Kruger system. There did, however, appear to be a weak trend for younger physicians to lean toward use of the Kruger morphology. The cutoff values for normal morphology preferred by clinicians in western U.S. practices tended to be higher, regardless of their preference for basic or Kruger assessment. For example, 70% of western clinicians preferring the basic morphology relied on a normal morphology cutoff of $50% compared with 54% of eastern clinicians. Likewise, 82% of western clinicians preferring the Kruger morphology relied on a normal morphology cutoff of $14% compared with 62% of eastern clinicians.
Postcoital Test We assessed details pertaining to physician practices concerning the PCT. Areas explored included preferred type of PCT (standard Sims-Huhner versus fractional), the length of abstinence before its performance, the determination of timing, whether the posterior vaginal fornix is aspirated for sperm, and the minimum motile sperm number for a normal PCT. Analysis of the PCT data revealed that 81.4% of all respondents normally order this test in the evaluation of the infertile couple. There was little variation in this trend by sex or geographic location of the clinician. Ninety percent of clinicians $50 years of age regularly ordered PCTs compared with 72% of younger physicians (P 5 .03), and 92% of private practitioners regularly ordered PCTs compared with 73% of their university-affiliated associates (P,.001). These significant differences remained after multivariate adjustment for all clinician characteristics. Among the selected population of clinicians who regularly ordered PCT, 89.7% reported that they relied on the standard Sims-Huhner test rather than the fractional PCT. Table 2 displays the data concerning the minimum motile sperm cutoff for a normal PCT and the methodology used to determine the optimum timing of the PCT. There was a significant trend of clinicians $50 years of age to select a greater minimum cutoff for a normal PCT in terms of sperm visualized per high-power field, whereas clinicians ,50 years of age gravitated toward allowing a lesser number of sperm present on the PCT to qualify it as a “normal” test (P 5 0.06). Regarding the test used to determine the optimal timing of the PCT, older physicians clearly relied more on cycle length estimation, whereas younger physicians relied FERTILITY & STERILITYt
TABLE 2 Percent of board-certified reproductive endocrinologists who order the PCT and use various methods for determining the timing of the PCT during the standard infertility evaluation by physician sex and age. Age of clinician (y)
Postcoital test characteristics Minimum cutoff for a normal PCT (sperm/HPF) $1 $3 $5 Cochran-Mantel-Haenszel x 2* Tests used to determine the timing of PCT Urine/blood LH Basal body temperature Cycle lengths Ultrasound to assess follicular growth Cochran-Mantel-Haenszel x 2*
$50
40–49
,40
No. (%)
No. (%)
No. (%)
15 (17.7) 42 (49.4) 28 (32.9)
26 (31.7) 14 (17.1) 35 (42.7) 7 (8.5)
55 (35.7) 25 (32.9) 78 (50.7) 40 (52.6) 21 (13.6) 11 (14.5) 14.4, P 5 .006
78 (52.4) 8 (5.4) 43 (28.9)
40 (53.3) 4 (5.3) 21 (28.0)
20 (13.4) 10 (13.3) 15.7, P 5 0.15
Note: HPF 5 high-power field. p Adjusted for type of practice (private, university), size of practice (,20, $20 patients), and geographic location (East, Midwest, West).
more often on urine or blood LH determinations (P 5 .015). Of clinicians who regularly ordered PCT assessment, 77% reported that they regularly aspirated the posterior vaginal fornix before the performance of the PCT. This distribution did not vary by clinician characteristics or practice location.
Hysterosalpingography The results of practice patterns involving the performance of the HSG are shown in Table 3. There was virtually no variation in the interpretation of the results by age or sex of the clinician. We observed that private practitioners more frequently performed HSG studies on their own without the assistance of a gynecologist or radiologist compared with university-affiliated colleagues. In addition, there was a significant geographic variability, with a greater proportion of physicians on the East Coast primarily performing their own HSG compared with practitioners in the West who tended to rely more heavily on their radiologist colleagues. Most practitioners on the East Coast and in the Midwest tended to use water-soluble media. However, although most practitioners in the western United States used water-soluble media, there was a significantly greater proportion of clinicians performing a water-soluble HSG followed by an oilbased dye “chaser” than that observed among physicians from both the Midwest and the East Coast. It is of interest that 70% of practitioners did not provide their patients with antibiotic prophylaxis before the performance of the HSG. However, physicians in private practice and those in the 265
TABLE 3 Percent of board-certified reproductive endocrinologists who order HSG and their methods for interpreting the results as part of the standard infertility evaluation by type of clinical practice and geographic region. Practice setting
HSG characteristics Clinician who generally performs HSG Self Colleague Radiologist Cochran-Mantel-Haenszel x 2* Contrast media used in performing HSG Oil-based media Water-soluble media Water-soluble followed by oil-based Cochran-Mantel-Haenszel x 2* Antibiotic prophylaxis in patients before HSG No Yes Cochran-Mantel-Haenszel x 2*
Practice location
University
Private
East
Midwest
West
No. (%)
No. (%)
No. (%)
No. (%)
No. (%)
123 (57.2) 138 (77.5) 43 (20.0) 4 (2.3) 49 (22.8) 36 (20.2) 28.3, P,.001
85 (64.4) 24 (18.2) 23 (17.4)
138 (75.0) 20 (10.9) 26 (14.1) 46.1, P,.001
38 (49.4) 3 (3.9) 36 (46.8)
14 (6.5) 15 (8.5) 183 (85.1) 146 (82.5) 18 (8.4) 16 (9.0) 0.33, P 5 .85
7 (5.3) 118 (89.4) 7 (5.3)
15 (8.2) 159 (86.4) 10 (5.4) 19.3, P 5 .001
7 (9.2) 52 (68.4) 17 (22.4)
160 (74.4) 114 (64.0) 55 (25.6) 64 (36.0) 3.1, P 5 .08
102 (77.3) 30 (22.7)
131 (71.2) 53 (28.8) 6.6, P 5 .04
41 (53.3) 36 (46.8)
p Adjusted for physician sex, age, and size of practice (,20, $20 patients).
western portion of the United States did tend to provide antibiotic coverage more often. Of those physicians who did prescribe prophylaxis before the HSG, virtually all chose a tetracycline-based regimen as their preferential choice. Only three clinicians chose a cephalosporin-based regimen, and two chose metronidazole-based regimens (data not shown).
Laparoscopy Overall, 64% of physicians chose to schedule laparoscopy during the follicular phase of the cycle (Table 4). This rate varied substantially by geographic region, with .80% of clinicians from the western United States preferring follicular phase laparoscopy and only half of the eastern physicians expressing the same preference. Most clinicians (67%) chose not to conduct hysteroscopy at the time of laparoscopy. However, all but 14 clinicians performed tubal lavage at the time of laparoscopy. Relative to university-based practitioners, there was a trend toward physicians in private practice to add a hysteroscopic evaluation in addition to laparoscopic assessment (P 5 .02). This trend pertained to a third of clinicians in private practice with only 17% of university-based clinicians offering the same prophylaxis. Of those clinicians who did provide regular antibiotic prophylaxis, 70% selected cephalosporin as their first choice with the remaining preferring a tetracycline-based regimen.
Immunologic Infertility Testing The final section of this survey inquired about miscellaneous tests and practices, including the assessment of immunologic infertility, the conditions under which a hamster egg sperm penetration assay is used, and the routine screen266
Glatstein et al.
Infertility evaluation practice patterns
ing of infertility patients for the luteinized, unruptured follicle syndrome. Approximately 72% of respondents stated they would conduct immunologic infertility testing under the appropriate circumstances with a somewhat greater proportion of private practitioners and western U.S. practitioners favoring this type of testing (Table 5). Of those practitioners who conduct immunologic testing, 95% stated that they would rely primarily on the immunobead method. Seven respondents said they would rely on ELISA testing, five on sperm agglutination, and two on a complement-dependent assay. Wide variation existed among clinicians who would or would not carry out hamster egg sperm penetration assays by practice setting and geographic location. With respect to using these assays for assessing abnormal semen analyses, 30% of private reproductive endocrinologists felt it was appropriate compared with 15% of university-based clinicians. Little difference was observed between private and university-based reproductive clinicians in the proportion of those who believed that these assays were appropriate for the assessment of unexplained infertility and as an IVF prognostic guide. In general, a significantly greater proportion of western U.S. clinicians believed that hamster egg sperm penetration assays was appropriate to use for assessing abnormal semen analyses, unexplained infertility, and as a prognostic IVF guide compared with both eastern and midwestern clinicians. As observed for the hamster egg sperm penetration assays, a significantly greater proportion of private practitioners (23.5%) reported that they routinely screen patients for Vol. 70, No. 2, August 1998
TABLE 4 Procedures followed by board-certified reproductive endocrinologists regarding laparoscopy as part of the standard infertility evaluation by type of clinical practice and geographic region. Practice setting
Laparoscopy characteristics When laparoscopy usually scheduled Follicular phase Luteal phase Scheduling availability Cochran-Mantel-Haenszel x 2* Hysteroscopy done at time of laparoscopy No Yes Cochran-Mantel-Haenszel x 2p Antibiotic prophylaxis in patients before laparoscopy No Yes Cochran-Mantel-Haenszel x 2*
Practice location
University
Private
East
Midwest
West
No. (%)
No. (%)
No. (%)
No. (%)
No. (%)
133 (61.6) 121 (67.6) 11 (5.1) 17 (9.5) 72 (33.3) 41 (22.9) 7.2, P 5 .03
70 (52.6) 10 (7.5) 53 (39.9)
122 (66.1) 15 (8.1) 48 (25.9) 19.3, P 5 .001
62 (80.5) 3 (3.9) 12 (15.6)
158 (72.8) 109 (60.6) 59 (27.2) 71 (39.4) 5.1, P 5 .02
87 (65.4) 46 (34.6)
127 (68.3) 59 (31.7) 0.29, P 5 .87
53 (68.0) 25 (32.0)
179 (82.5) 119 (66.1) 38 (17.5) 61 (33.9) 13.9, P,.001
106 (79.7) 27 (20.3)
136 (73.1) 50 (26.9) 3.6, P 5 .17
56 (71.8) 22 (28.2)
p Adjusted for physician sex, age, and size of practice (,20, $20 patients).
luteinized unruptured follicles compared with only 8% of university-based clinicians. Although western U.S. practitioners did screen for this syndrome more often than eastern or midwestern practitioners, the difference was substantially less than that observed for hamster egg penetration assays.
DISCUSSION Our first area of questioning concerned semen analysis. It appears that physicians in a private practice setting have a significantly higher rate of recommending to patients that the semen analysis be produced after $2 days of abstention. This was not seen in all the physicians and those in a university practice. It is of interest that the recommendations put forth by the World Health Organization (WHO) state that “ideally the sample should be collected after a minimum of 48 hours but not longer than seven days of sexual abstinence” (2). It is unclear why younger physicians and those in private practice would be more strictly adherent to these guidelines, but this may be a result of fellowship training practices, which only recently have been introduced.
morphological examination, i.e., the basic (WHO) versus the Kruger strict determination of morphology. Although 66% of all respondents stated that they did favor the basic morphology, there was a statistically significant trend among younger physicians to favor the Kruger morphology. Of interest is a recent study by Morgentaler et al. (3) that directly compared the WHO and strict criteria methodologies using IVF outcome as an endpoint. In this study the traditional basic morphology demonstrated a higher sensitivity and negative predictive value compared with strict criteria. We would suggest that the modern evaluation of sperm is an evolving process, and given the rapid developments of microfertilization and intracytoplasmic sperm injection (ICSI), need to be reexamined and reevaluated in light of these new technologies.
Overall, approximately 60% of physicians prefer the manual count when performing an evaluation of a semen sample compared with a computer-assisted analysis. There was, however, no discernible difference between the various identifiable individual physician groups. The use of the computer-assisted semen analysis is an evolving field, and as the methods are perfected and equipment becomes more affordable and readily available, we may expect to see practice patterns shift in this regard.
The PCT is an area of infertility testing that has engendered much controversy in recent years. For example, perhaps one of the most vexing questions surrounding its use is the cutoff that defines a normal test. This number has varied in studies from a high of 20 sperm per high-power field to as low as one moving spermatozoa (4). Our data demonstrate that although older physicians tend to have a higher threshold for determining a satisfactory PCT, younger physicians tend to allow fewer numbers of sperm to be present to qualify it as a normal test. Likewise, younger physicians tended to rely on more modern methods of ovulation detection, such as urine or blood LH assessment to determine the timing of the PCT, versus older physicians who tended to rely more heavily on traditional methods of estimating ovulation such as cycle lengths.
Another unsettled topic of semen analysis is the type of
These differences may reflect simply the phenomenon of
FERTILITY & STERILITYt
267
TABLE 5 Percent of board-certified reproductive endocrinologists who test for immunologic infertility, conditions under which practitioners use hamster egg sperm penetration, and whether practitioners screen infertility patients for luteinized unruptured follicle syndrome by type of clinical practice and geographic region. Practice setting
Other infertility tests and procedures Immunologic infertility testing Would not test Would test Cochran-Mantel-Haenszel x 2* Conditions under which hamster egg sperm penetration assays are used Abnormal semen tests Would not test Would test Cochran-Mantel-Haenszel x 2* Unexplained infertility Would not test Would test Cochran-Mantel-Haenszel x 2* Prognostic IVF guide Would not test Would test Cochran-Mantel-Haenszel x 2* Clinicians that routinely screen for luteinized unruptured follicle syndrome Would not screen Would screen Cochran-Mantel-Haenszel x 2*
Practice location
University
Private
East
Midwest
West
No. (%)
No. (%)
No. (%)
No. (%)
No. (%)
73 (33.6) 39 (21.6) 144 (66.4) 142 (78.4) 4.5, P 5 .04
48 (36.1) 85 (63.9)
48 (25.8) 138 (74.2) 8.0, P 5 .02
16 (20.5) 62 (79.5)
182 (84.7) 127 (70.2) 33 (15.4) 54 (29.8) 12.2, P,.001
113 (85.0) 20 (15.0)
151 (82.1) 33 (17.9) 23.8, P,.001
45 (57.7) 33 (42.3)
141 (65.6) 105 (58.0) 74 (34.4) 76 (42.0) 2.9, P 5 .08
98 (73.7) 35 (26.3)
122 (66.3) 62 (33.7) 35.0, P,.001
26 (33.3) 52 (66.7)
153 (71.2) 123 (68.0) 62 (28.8) 58 (32.0) 0.28, P 5 .6
111 (83.5) 22 (16.5)
122 (66.3) 62 (33.7) 20.8, P,.001
42 (53.9) 36 (46.1)
199 (92.1) 137 (76.5) 17 (7.9) 42 (23.5) 18.0, P,.001
122 (92.4) 10 (7.6)
149 (81.0) 35 (19.0) 5.8, P 5 .06
64 (82.1) 14 (18.0)
* Adjusted for physician sex, age, and size of practice (,20, $20 patients).
younger, more recently trained physicians, using more recently introduced technology. This is a concept that has been well-documented in many areas of medicine including data that we previously reported (1). The subject of oil- versus water-soluble contrast media in the performance of the hysterosalpingogram has long been a matter of debate in the reproductive endocrinology literature. There are proponents of both approaches, with those who favor oil contrast quoting literature demonstrating that oilsoluble contrast media has a therapeutic effect in regards to pregnancy rates compared with water-soluble media. A metaanalysis by Watson and coworkers (5) encompassing four randomized clinical trials and six nonrandomized controlled studies evaluated pregnancy rates after the use of each of these contrast media. They concluded that the odds ratio for pregnancy was 1.9 for the oil-based dye compared with the use of water-soluble contrast media. In addition to this important increase in the probability of pregnancy with oil media, proponents of the oil method also claim that the incidence of abdominal pain, vaginal bleeding, and pelvic inflammatory disease (PID) is significantly lower compared with that observed in women experiencing watersoluble media. Loy et al. (6) found that the sensitivity and predictive value of HSG was higher in oil-soluble media 268
Glatstein et al.
Infertility evaluation practice patterns
compared with water-soluble media when using a laparoscopic diagnosis as the gold standard. Opponents of oilbased contrast media cite the danger of a potential increased risk of embolism if intravasation occurs, and the concern that if oil is trapped within a hydrosalpinx, a granuloma may be formed. However, with modern fluoroscopic methods of performing the HSG, the risk of embolism virtually is eliminated. In addition, the issue of granuloma formation of oil media within tubes is somewhat theoretical. With this in mind, it is somewhat difficult to explain why most physicians choose to continue to use water-soluble media, given the relative advantages of the oil method. At the very minimum, a water-soluble hysterosalpingogram, followed by an oil “chaser” if the tubes are patent, would provide the advantages of oil without the disadvantages of potential trapping of oily fluid within a hydrosalpinx. It may simply be that the metaanalysis study, which came out of the United Kingdom, has not been disseminated widely within training programs and the reproductive endocrine community. Concerning antibiotic prophylaxis, our data show that most physicians do not provide antibiotic prophylaxis routinely for HSG. Although the literature concerning the use of antibiotic prophylaxis and HSG is somewhat scarce, it would Vol. 70, No. 2, August 1998
seem logical that, given the devastating consequences of PID after a hysterosalpingogram, it is certainly not unreasonable to recommend antibiotic prophylaxis, given the relatively low morbidity associated with a short course of prophylaxis. A recently published bulletin by the American College of Obstetricians and Gynecologists likewise recommends antibiotic coverage in all patients undergoing an HSG procedure (7). Laparoscopy generally is considered a basic element of the infertility workup, although its routine use in cases in which there is no compelling evidence for pelvic disease and in which the hysterosalpingogram is normal, may not be cost-effective. Most physicians schedule the laparoscopy during the follicular phase of the cycle. This is logical because there is less of a chance of inadvertently performing a laparoscopy during pregnancy. Although most physicians who were surveyed do not perform a hysteroscopy routinely at the time of laparoscopy, there was a trend for physicians in private practice to offer this procedure compared with their university-based colleagues. Immunologic testing for infertility is a controversial topic. It is of interest that physicians on the West Coast tended to order more immunologic testing than their East Coast colleagues. This may be reflective of fellowship and infertility training practices on the West Coast. Except for physicians residing on the West Coast, when encountering patients with unexplained infertility, most physicians on the East Coast and in the Midwest stated that they would not routinely order hamster egg sperm penetration assay. A study using life-table analyses and monthly fecundity rates by O’Shea et al. (8) has shown that the hamster egg penetration score is not predictive of the incidence of pregnancy nor of time to conception. It would appear that, especially with the more prevalent use of IVF and ICSI for male factor, the hamster egg sperm penetration assay generally has fallen out of favor with reproductive endocrinologists. The final category of infertility testing concerns routine screening for the luteinized, unruptured follicle syndrome. Most clinicians stated that they would not screen for this condition. We may speculate that the reason for this lack of enthusiasm is the controversy over whether this entity actually exists and how one tests for it. Is the diagnosis made with laparoscopy with visualization of ovarian stigma, progesterone determinations, or the lack of follicular reduction
FERTILITY & STERILITYt
in size on ultrasound? (9, 10). Until this condition is defined more clearly and rigid criteria for its diagnosis are determined, it will remain most likely a marginal and elusive entity at best. In summary, we found that there were distinct practice pattern differences among infertility practitioners, especially with regards to age and geographic location. Younger, board-certified, more recently trained physicians tended to incorporate more modern methodologies into their infertility practice. Data such as these regarding the actual infertility workup as practiced by board-certified specialists is an important first step in seeking to better use resources in the treatment of infertile couples. These results may assist those who may seek to develop national standards in the evaluation of infertility. It will be interesting to observe how these practice patterns will change over time as new technologies become integrated into the procedures available to infertility practitioners.
Acknowledgment: The authors thank Ms. Paula Ouellette for assistance in the preparation of this manuscript.
References 1. Glatstein IZ, Harlow BL, Hornstein MD. Practice patterns among reproductive endocrinologists: the infertility evaluation. Fertil Steril 1997;67:443–51. 2. World Health Organization. Laboratory manual for the examination of human semen and semen-cervical mucus interaction. 3rd ed. New York: Cambridge University Press, 1992:3. 3. Morgentaler A, Fung MY, Harris DH, Powers RD, Alper MM. Sperm morphology and in-vitro fertilization outcome: a direct comparison of World Health Organization and strict criteria methodologies. Fertil Steril 1995;64:1177– 82. 4. Griffith CS, Grimes DA. The validity of the post-coital test. Am J Obstet Gynecol 1990;162:615–20. 5. Watson A, Vandekerchhove P, Lilford R, Vail A, Brosens I, Hughes E. A meta-analysis of the therapeutic role of oil-soluble contrast media at hysterosalpingography: a surprising result? Fertil Steril 1994;61:470 –7. 6. Loy RA, Weinstein FG, Seibel MM. Hysterosalpingography in perspective: the predictive value of oil-soluble versus water-soluble contrast media. Fertil Steril 1989;51:170 –2. 7. The American College of Obstetricians and Gynecologists. Antibiotics and gynecologic infections. ACOG Educational Bulletin No 237, June 1997. 8. O’Shea DL, Odem RR, Cholewa C, Gast MJ. Long term follow-up of couples after hamster egg penetration testing. Fertil Steril 1993;60: 1040 –5. 9. Hamilton MPR, Fleming R, Coutts JRT, Macaughton MC, Whitfield CR. Luteal cysts and unexplained infertility: biochemical and ultrasonic evaluation. Fertil Steril 1990;50:32–7. 10. Scheenges E, Velde ER, Kremer J. Inspection of the ovaries and steroids in serum and peritoneal fluid at various time intervals after ovulation in fertile women: implications for the luteinized unruptured follicle syndrome. Fertil Steril 1990;54:38 – 41.
269