Ultrasonographic monitoring of follicular growth for luteal phase defects*

FERTILITY AND STERILITY Copyright © 1987 The American Fertility Society

Vol. 48, No.3, September 1987 Printed in U.S.A.

Ultrasonographic monitoring of follicular growth for luteal phase defects*

Yu-Kang Ying, M.D.t:J: Douglas C. Daly, M.D.§ John F. Randolph, M.D.:j: Carlos E. Soto-Albors, M.D.II:j:

Donald B. Maier, M.D.:j: Cecilia L. Schmidt, M.D.§:j: Daniel H. Riddick, M.D., Ph.D.**:j:

Department of Obstetrics and Gynecology, University of Connecticut Health Center, Farmington, Connecticut, and University of Massachusetts, Worcester, Massachusetts

In the evaluation of 39 patients with untreated and treated luteal phase defect (LPD), serial ultrasonographic monitoring of follicular development identified three morphologically distinct growth patterns: normal-sized follicles, small follicles, and luteinizedunruptured follicles. All three patterns were observed in both untreated (46% had normal-sized follicles, 39% had small follicles, 15% had luteinized-unruptured follicles) and treated patients. A small follicle was observed uncommonly in histologically corrected LPD patients (6%). However, a luteinized-unruptured follicle (38%) may persisit or be induced in situations where clomiphene citrate has been used to correct LPD or induce ovulation. Ultrasonographic evaluations of follicular growth in luteal phase defect support the theory that luteal phase defect represents a spectrum of normal and abnormal ovarian cycle events. Fertil Steril 48:433, 1987

Luteal phase defect (LPD) represents a spectrum of dysfunction that may be reflected by both endometrial and ovarian cycle abnormalities. Nu-

Received March 27, 1986; revised and accepted May 1, 1987.

* Presented in part at the forty-first annual meeting of the American Fertility Society, Chicago, Illinois, September 28 to October 2, 1985. :j: Department of Obstetrics and Gynecology, University of Connecticut Health Center. § Department of Obstetrics and Gynecology, University of Massachusetts. t:j: Present address and reprint requests: Yu-Kang Ying, M.D., Department of Obstetrics and Gynecology, Nassau' County Medical Center, 2201 Hempstead Turnpike, East Meadow, New York 11554. § Present address: University of Massachusetts Medical School, Worcester, Massachusetts. II Present address: David Grant United States Air Force Medical Center, Department of Obstetrics and Gynecology, Travis AFB, California. § Present address: University of Medicine and Dentistry of New Jersey, New Jersey Medical School, Newark, New Jersey. ** Present address: University of Vermont, Department of Obstetrics/Gynecology Burlington, Vermont. Vol. 48, No.3, September 1987

merous studies have appeared in the literature describing this clinical entity.1 The timed endometrial biopsy as a diagnostic method has been validated by clinical data showing its predictive value for therapeutic outcomes during the treatment of LPD. 2 Correction of luteal phase defect, as indicated by a corrected timed endometrial biopsy, was associated with improved fertility, whereas failure to correct the histology was associated with a poor prognosis for viable pregnancy.2 However, except in cases of hyperprolactinemia or hypothyroidism where specific treatments are available, the initial choice of treatment is empiric. The adequacy of therapy can only be assessed by repeat endometrial biopsy or a 6-month or longer therapeutic trial. Based on this protocol, if the biopsy is uncorrected, the next form of therapy would be initiated in the following cycle. Even though this approach allows improved fertility, multiple biopsies frequently are necessary. Since the real-time ultrasound sector scanner capable of monitoring follicular growth became available, normal follicular growth patterns have been Ying et al.

Luteal phase defect and ultrasound

433

described. The mean follicular diameter in normal spontaneous cycles before ovum release varied in different studies from 21.1 to 24.6 mm, with a range of 17 to 29 mm.3-5 Ultrasonographic evaluations also allow detection of abnormal follicular growth, such as the luteinized-un ruptured follicle syndrome. 5,6 The present study was conducted to observe morphologic features of follicular growth in LPD patients as observed by real-time ultrasonographic evaluations. It was hoped that ultrasonographic monitoring of follicular growth would identify subsets of LPD patients with follicular phase-ovarian cycle abnormalities. This might eventually provide a guide to therapy and reduce the number of endometrial biopsies required during therapeutic trials. MATERIALS AND METHODS

Twenty-six patients diagnosed to have LPD between March 1984 and June 1985 in the Unive'rsity of Connecticut or the University of Massachusetts Reproductive Endocrinology clinics were studied prospectively for follicular dynamics. The patients underwent serial pelvic ultrasonographic evaluations after the diagnosis of LPD and prior to the institution of therapy. All patients had biphasic basal body temperatures (BBTs) and normal postcoital tests. One patient was receiving bromocriptine for hyperprolactinemia. The diagnosis of LPD was established by abnormal endometrial biopsies on days 11 to 13 of the luteal phase in two cycles, as dated by established criteria7 or by luteal phase lengths of 10 days or less. Each sample was reviewed histologically, dated, and compared with both presumed ovulation by BBT and subsequent menstruation. When histologic date lagged for 2 or more days, the biopsy was considered out of phase. Two patients in this group had LPD based on luteal phase lengths. Ultrasound scans (USs) were performed at the University of Connecticut with a General Electric Company RT-3000 Scanner GE-SRT (General Electric Company, Medical Systems Group, Milwaukee, WI) at 3.5 MHz by authors other than Dr. Daly. This machine previously had been standardized for size of follicle the day prior to rupture as 18 s rupture s 25, with a mean daily growth of 2.35 mm. At the University of Massachusetts, an ADR 4000 LIS Scanner (Advanced Technology Laboratories, Inc., Bellevue, W A) at 3.5 MHz was used by Dr. Daly. This machine had been standardized for size of follicle prior to rupture of 17 s rupture 434

Ying et al.

Luteal phase defect and ultrasound

s 23.7, with a mean of 20.7 mm and average daily growth of 2.1 mm in 150 spontaneous ovulatory cycles. The initial US was performed 3 to 4 days before the expected day of ovulation and repeated every 1 to 2 days, depending on follicle size, until rupture had occurred or until 3 to 4 days of temperature elevation on BBT. The diagnosis of follicular rupture was made by decrease in size or collapse of the follicular cyst with the development of internal echoes and presence of cul-de-sac fluid. Luteinized-unruptured follicle syndrome (LUFS) was diagnosed when a follicle failed to rupture by these criteria and the cyst wall demonstrated gradual thickening during 3 to 4 days of temperature elevation by BBT. Using the standards of each machine, the diagnosis of a small follicle was made when the follicle size on the day prior to rupture was less than the lower limit of normal. When the last scan was 2 days prior to documentation of rupture and demonstrated a follicle 3 mm or smaller than the lower limit of normal (follicle growth on these machines averages 2.35 mm and 2.1 mm, respectively), this also would be diagnosed as a small follicle. When cycles could not be defined clearly as normal or abnormal for any reason, they were repeated. The incidence of LUFS and rupture at small follicular size in this study was compared by chi-square analysis to a similar previously published group of patients with unexplained infertility and normal late luteal phase endometrial biopsies. 6 Statistical significance was defined as P < 0.05 (Table 1). Seventeen patients previously diagnosed and treated for LPD who had failed to conceive after 6 months of adequate treatment (determined by a corrected endometrial biopsy) underwent serial US for follicular dynamics. They were assessed for ovulatory quality by the same criteria. Seven patients had LPD based on short luteal phase. Seven patients underwent diagnostic laparoscopies where endometriosis was found in four (three had nor-

Table 1 Relative Frequencies of Normal and Abnormal Follicles in Untreated LPD Patients and Unexplained Infertile Patients a Normal follicle

Small follicle or lutenized unruptured follicle

12

14 6

LPD Unexplained infertility a

X2 , 7.40;

27

df, 1; P < 0.01. Fertility and Sterility

Table 2 Relative Frequencies of Abnormal Follicles in Luteal Phase Defecta Small follicle Untreated Treated a

p

=

Lutenized unruptured follicle

10

4

1

5

0.0378 by Fisher's exact test.

mal-sized follicle, one had LUFS). These patients were excluded from our study. Among the remaining 13 patients, 5 were being treated with progesterone suppositories and eight with clomiphene citrate. The relative frequency of abnormal follicles in the untreated prospective group was compared with the histologically corrected retrospective group, as shown in Table 2. The difference was evaluated by Fisher's exact test and statistical significance was defined at P < 0.05. RESULTS

Among 26 patients in the prospectively evaluated group, 12 patients ruptured normal follicles (46%), 10 ruptured small follicles (39%), and 4 demonstrated LUFS (15%). Compared with patients with unexplained infertility (33 patients, 27 normal, 3 small follicles, 3 LUFS),6 this represented a significant difference (x 2 = 7.36, P < 0.01), primarily due to the increase in small follicles at the time of rupture (Table 1). In the retrospective group of 13 patients, 7 patients ruptured normal-sized follicles, 1 (6%) patient ruptured a small follicle and 5 patients demonstrated LUFS (Table 2). Of the 5 patients demonstrating LUFS, 4 patients were receiving clomiphene citrate for the treatment of LPD. As shown in Table 2, significantly fewer small follicles were found in the treated retrospected group compared with the untreated prospective group. DISCUSSION

In the evaluation of patients with LPD, serial ultrasonographic monitoring of follicular development identified three growth patterns, normalsized rupturing follicles, small rupturing follicles, and LUFS. These findings were similar to those reported by Check et al. 8 Patients with LPD demonstrate an increased incidence of abnormal folliVol. 48, No.3, September 1987

culogenesis, primarily due to a high incidence of follicles rupturing at a small size: 39% of LPD patients versus 9% for patients with unexplained infertility. The frequency of this abnormality differs between untreated and treated LPD patients. Compared with untreated LPD, histologically corrected LPD was associated with significantly fewer small follicles. In patients who fail to conceive on adequate therapy as determined by corrected endometrial biopsy, the primary follicular abnormality is LUFS: 9% of patients with unexplained infertility versus 15% of prospective LPD patients versus 38% in patients with corrected LPD but not conceiving. This latter group was primarily composed of patients on clomiphene citrate. Unfortunately, we do not know what the original follicular dynamics of these patients were. It is apparent that patients with LUFS/LPD will not respond to progesterone replacement. Since this represents 15% to 38% of the LPD patients, serial US is important prior to initiating therapy. The high incidences of LUFS in patients on clomiphene citrate, even with corrected biopsies, warrant repeat of the serial US before treatment is presumed to be adequate. Since only half of LPD patients respond to progesterone, it is also apparent that some of the small follicles at rupture or normal follicles at rupture must also be progesterone nonresponders. Unfortunately, a predictable correlation between follicle size and success on progesterone suppositories has not been documented. Small follicles «18 mm) were not associated with pregnancy in the donor insemination study with US correlation. 9 However, these patients were not screened for LPD by biopsy in these cycles and this could be the reason for the failure to conceive. An alternative theory would suggest that the smaller follicles might be less mature and that the ovum might not resume normal meiosis. The cycle would thereby be infertile regardless of the state of the endometrium. This study does not address these questions. It is clear, however, that the incidence of follicle rupture at small size is much more common in LPD cycles than in patients with unexplained infertility. In conclusion, serial US for follicular dynamics indicates that many LPD patients have abnormal folliculogenesis. Patients with a LUFS pattern require treatment with ovulation-inducing agents, and LUFS seem to occur occasionally with clomiphene therapy. The optimal therapy for patients with LPD and follicular rupture at small size, however, is yet to be determined. Ying et al.

Luteal phase defect and ultrasound

435

I

REFERENCES 1. Andrews WC: Luteal phase defects. Fertil Steril 32:501,

1979 2. Daly DC, Walters CA, Soto-Albors CE, Riddick DH: Endometrial biopsy during treatment of luteal phase defects is predictive of therapeutic outcome. Fertil Steril 40:305, 1983 3. Bryce RL, Shuter B, Sinosich MJ, Stiel IN, Picker RH, Saunders DM: The value of ultrasound, gonadotropin and estradiol measurements for precise ovulation prediction. Fertil Steril 37:42, 1982 4. Kerin JF, Edmonds DK, Warnes GM, Cox LW, Seamark RF, Mathews CD, Young GB, Baird DT: Morphological and functional relations of Graafian follicle growth to ovulation in women using ultrasonic, laparoscopic, and biochemical measurements. Br J Obstet Gynaecol 88:81, 1981 5. O'Herlihy C, de Crespigny LC, Lopata A, Johnston I, Hoult

436

Ying et al. Luteal phase defect and ultrasound

6.

7. 8.

9.

I, Robinson H: Preovulatory follicular size: comparison of ultrasound and laparoscopic measurements. Fertil Steril 34:24,1980 Daly DC, Soto-Albors CE, Walters CA, Ying Y-K, Riddick DH: Ultrasonographic assessment of luteinized unruptured follicle syndrome in unexplained infertility. Fertil Steril 43:62,1985 Noyes RW, Hertig AI, Rock J: Dating endometrial biopsy. Fertil Steril 1:3, 1950 Check JH, Goldberg BB, Kurtz A, Adelson HG, Rankin A: Pelvic sonography to help determine the appropriate therapy for luteal phase defects. Int J Fertil 29:156, 1984 Marinho AO, Sallam HN, Goessens LKV, Collins WP, Rodeck CH, Campbell S: Real time pelvic ultrasonography during the periovulatory period of patients attending an artificial insemination clinic. Fertil Steril 37:633, 1982

Fertility and Sterility