FERTILITY AND STERILITY
Vol. 53, No.6, June 1990
Copyright© 1990 The American Fertility Society
Printed on acid-free paper in U.S.A.
Failure of body mass index or body weight to influence markedly the response to ovarian hyperstimulation in normal cycling women
Claire G. Lewis, M.D. Graham M. Warnes, Ph.D. Xinjung Wang, M.Ag.Sci. Colin D. Matthews, F.R.A.C.O.G.* Reproductive Medicine Unit, The University of Adelaide, The Queen Elizabeth Hospital, Woodville, South Australia
In vitro fertilization (IVF) or gamete intrafallopian transfer (GIFT) programs require controlled ovarian hyperstimulation to optimize the chance of pregnancy. One standard regime for hyperstimulation utilizes clomiphene citrate (CC, Clomid; Merrell Dow, Sydney, Australia), human menopausal gonadotropin (hMG, Humegon; Organon, Melbourne, Australia, or Pergonal; Serono, Melbourne, Australia, distributed by CSL, Melbourne, Australia), and human chorionic gonadotropin (hCG, Profasi; Serono, distributed by CSL). However, the response to ovarian hyperstimulation is frequently variable in terms of hMG requirements and/or the number of days required to reach a satisfactory response; in addition, a minority of subjects fail to respond adequately. The reasons for this variability are frequently unclear, but body weight of the subjects is one variable to be considered. Excess body weight is known to impair the response to CC, and it has been suggested that obesity per se is associated with an altered pituitary response to endogenous luteinizing hormone (LH)releasing hormone from the hypothalamus. Conversely, subjects with severe weight loss commonly show a disturbance in the hypothalamic-pituitarygonadal axis and the loss of CC responsiveness. 1 In
Received September 6, 1989; revised and accepted February 16,1990. * Reprint requests: Colin D. Matthews, F.R.A.C.O.G., Reproductive Medicine Unit, The University of Adelaide, The Queen Elizabeth Hospital, Woodville 5011, South Australia. Vol. 53, No.6, June 1990
normally cycling women, Halme et al. 2 have demonstrated an increased response to hMG in women with lower body weight so that the day of oocyte collection was earlier than in normal weight subjects with an adverse effect on the frequency of oocyte .fertilization. This retrospective study has analyzed the response to a standardized treatment regime for controlled ovarian hyperstimulation associated with IVF and GIFT with particular reference to the variation in the body weight of subjects. In some contrast to the above studies, the results failed to confirm a major influence of body weight parameters on the ovarian response in these reproductively normal women. MATERIALS AND METHODS
Three hundred sixty-eight subjects referred to the Reproductive Medicine Unit at The Queen Elizabeth Hospital for IVF and GIFT between 1985 and 1988 were included in this retrospective study. All of the selected women had regular reproductive cycles and underwent a standard regime of controlled ovarian hyperstimulation. Patients with severe endometriosis or who had a single ovary which might have precluded a normal ovarian response were excluded, and only cycles of women with a currently known height and weight having their initial cycle of treatment were included. The standard regime of ovarian hyperstimulation used for all subjects in this study utilized CC 100 mg days 5 to 9 and hMG 150 IU (Pergonal, Lewis et al.
Communications-in-brief
1097
Table 1
Relationship Between BMI and Ovarian Response a No. of subjects
Days of treatment
Maximum
BMiindex
27.6
34 114 72 112 36
12.2 ± 1.0 12.6 ± 1.0 12.8± 1.0 13.0 ± 1.0 13.1 ± 1.0
8.9 ± 4.1 9.3 ±4.5 9.7±4.8 8.0±4.2 7.1 ± 3.5
E2
Oocytes recovered
Oocytes fertilized
6.4 ± 3.2 6 6.6±4.0 7.0±4.3 6.4 ± 3.9 4.8 ± 2.6 6
4.7 ± 2.3 5.1 ± 3.3 5.6± 4.4 4.7±.3.8 3.7 ± 2.4
nM/L
a
Values are means± 2 SD.
Serono or Humegon, Organon) days 6, 8, and 10. The ovarian response was carefully monitored on a day-to-day basis, commencing on cycle day 10 by the measurement of plasma estradiol (E 2 ), progesterone, LH, and by abdominal (to 1986) and vaginal ultrasonography of the ovarian follicle(s). Additional daily injections ofhMG (150 IU) were continued until the leading follicles were ::::;15 mm (mean diameter). For the analysis, the number of ampules of hMG required, the peak serum E 2 , the day of hCG or LH surge, and the number of oocytes collected, inseminated, and fertilized were all recorded together with the number of pregnancies achieved. Body mass index (BMI) values for individual women were calculated by the ratio of body weight (kg) divided by the height2 (m 2 ). Total body weight for the group of subjects ranged between 41 and 106 kg (mean ± SD, 61.4 ± 11.5). Height ranged between 148 and 182 em (163 ± 8.2). Regression analysis was not used because the samples had a skewed distribution even after Log transformation. For the purpose of comparison, the sample population was split into five groups according to BMI: underweight ( 90th percentile). Statistical analysis was performed by comparing the median of the groups using MannWhitney nonparametric methods.
6
Significantly different (P < 0.05).
Whereas a trend to increasing E 2 maximum values was identified with both increased number ofhMG ampules used and the increased days of treatment, no significant relationships were found. The data were therefore treated as being homogenous. The relationship between BMI and the ovarian response to hyperstimulation as determined by the serum E 2 level before the day of hCG is shown in Table 1. Whereas the results indicated a tendency for the number of days of treatment required before hCG to increase in subjects with an increased BMI, the difference between the two extreme groups was 27.6 group, the difference was not significant. The mean number of oocytes recovered was greatest in the middle of the BMI range (7.0). The mean number in the BMI < 19.1 group (6.4) was higher (P < 0.05) than in the BMI > 27.6 group (4.8). No differences were apparent in the mean number of oocytes fertilized between groups. Analysis of the data using total body weight in place of BMI demonstrated no significant differences in any parameter examined. A total of 96 pregnancies were established from the IVF and GIFT procedures. The rate of occurrence of clinical pregnancy in the five groups of subjects was 30%, 21%, 24%, 26%, and 20%, respectively. No statistically significant differences between groups were evident.
RESULTS
Given the variation in the hMG dosage used per cycle (range 450 to 1,200 IU) and/or the varying number of days of treatment (range 3 to 9), the data were analyzed with respect to these factors. For the majority of subjects, hCG was given on day 12 (n = 121), day 13 (n = 125), or day 14 (n = 61). 1098
Lewis et al.
Communications-in-brief
DISCUSSION
Weight and body mass are known to be of some importance to the maintenance of regular reproductive cycles. Certainly, either excess weight or marked weight loss can be associated with cycle disturbance. A direct influence of body weight on Fertility and Sterility
ovarian response to exogenous gonadotropins has been shown for subjects requiring induction of ovulation, with obese women requiring larger doses of hMG 3 or CC 4 to effect an optimal response. Data relating to the use of CC and hMG for ovarian hyperstimulation in reproductively normal women are sparse, but Halme et al. 2 found theresponse to hMG was related to body weight, noting particularly the rapid and the hyper-response of underweight women, but with associated poorer oocyte fertilization rates. These authors showed a proven benefit of a reduced dosage schedule over an extended time frame. 5 The number of days of treatment required to reach satisfactory pre-hCG E 2 levels in this study were similar in underweight, normal weight, and obese women. The results thus failed to confirm the earlier findings, although the stimulation regime used differed in that our subjects used CC from days 5 to 9 with hMG supplementation on days 6, 8, and 10, in contrast to the regime of Halme et al., 2 which utilized 150 IU hMG from day 3. It may be, therefore, that the time of commencement of the stimulation regime and perhaps the choice of drugs are critical factors to consider with respect to the effect of body weight or BMI on ovarian responsiveness. It should be also recognized that since a combination of CC and hMG was utilized, the assessment as to whether either of the agents used are weight dependent is difficult. The results of this study may not therefore be applicable to all ovarian hyperstimulation regimes. Whereas the numbers of oocytes recovered were somewhat greater in the underweight than the obese women, these were not different from normal weight subjects. Fertilization rates were comparable in all five groups. When body weight was considered in contrast to BMI, no statistical significance could be derived between any of the parameters measured. It is possible that treating women with normal reproductive function with CC and supraphysiological doses of hMG may well override subtle differences in the levels of hypothalamic pituitary
Vol. 53, No.6, June 1990
ovarian sensitivity. In addition, the large doses used may overcome variability in intramuscular absorption. Drug absorption and metabolic clearance studies related to body weight would be helpful to clarify these possibilities. SUMMARY
A retrospective analysis was performed of 368 normally cycling women treated with a single cycle of a standard ovarian hyperstimulation regime (CC 100 mg days 5 to 9 and hMG 150 IU days 6, 8, and 10) associated with either an IVF or GIFT program. Neither the peak serum E 2 level attained nor the number of days of stimulation required bore a relationship to the BMI or the total body weight of these women. Whereas the mean number of oocytes aspirated from women with BMI < 19.1 was higher (6.4 ± 3.2) compared with obese women (BMI > 27.6, 4.8 ± 2.6), the rate of fertilization was not different for both BMI extremes. It is concluded that factors other than BMI or total body weight have more important influences on the response to hyperstimulation in normal women. Ackrwwledgments. To Barbara Godfrey, M.A.(Oxon), for maintenance of the data base and to the clinical and nursing staff of the Reproductive Medicine Programs for clinical management of the subjects. REFERENCES 1. Reid RL, Van Vugt DA: Weight-related changes in repro-
ductive function. Fertil Steril48:905, 1987 2. Halme J, Hammond MG, Talbert LM, O'Rand M, Bailey L, Sloan C: Positive correlation between body weight, length of human menopausal gonadotropin stimulation, and oocyte fertilization rate. Fertil Steril45:372, 1986 3. Chong AP, Rafael RW, Forte CC: Influence of weight in the induction of ovulation with human menopausal gonadotropin and human chorionic gonadotropin. Fertil Steril46:599, 1986 4. Shepard MK, Balmaceda JP, Leija CG: Relationship of weight to successful induction of ovulation with clomiphene citrate. Fertil Steril32:641, 1979 5. Halme J, Hammond MG, Bailey L, Talbert LM: Lower doses of human menopausal gonadotropin are associated with improved success with in vitro fertilization in women with low body weight. Am J Obstet Gynecol158:64, 1988
Lewis et al.
Communications-in-brief
1099
Vol. 53, No.6, June 1990
Copyright© 1990 The American Fertility Society
Printed on acid-free paper in U.S.A.
Failure of body mass index or body weight to influence markedly the response to ovarian hyperstimulation in normal cycling women
Claire G. Lewis, M.D. Graham M. Warnes, Ph.D. Xinjung Wang, M.Ag.Sci. Colin D. Matthews, F.R.A.C.O.G.* Reproductive Medicine Unit, The University of Adelaide, The Queen Elizabeth Hospital, Woodville, South Australia
In vitro fertilization (IVF) or gamete intrafallopian transfer (GIFT) programs require controlled ovarian hyperstimulation to optimize the chance of pregnancy. One standard regime for hyperstimulation utilizes clomiphene citrate (CC, Clomid; Merrell Dow, Sydney, Australia), human menopausal gonadotropin (hMG, Humegon; Organon, Melbourne, Australia, or Pergonal; Serono, Melbourne, Australia, distributed by CSL, Melbourne, Australia), and human chorionic gonadotropin (hCG, Profasi; Serono, distributed by CSL). However, the response to ovarian hyperstimulation is frequently variable in terms of hMG requirements and/or the number of days required to reach a satisfactory response; in addition, a minority of subjects fail to respond adequately. The reasons for this variability are frequently unclear, but body weight of the subjects is one variable to be considered. Excess body weight is known to impair the response to CC, and it has been suggested that obesity per se is associated with an altered pituitary response to endogenous luteinizing hormone (LH)releasing hormone from the hypothalamus. Conversely, subjects with severe weight loss commonly show a disturbance in the hypothalamic-pituitarygonadal axis and the loss of CC responsiveness. 1 In
Received September 6, 1989; revised and accepted February 16,1990. * Reprint requests: Colin D. Matthews, F.R.A.C.O.G., Reproductive Medicine Unit, The University of Adelaide, The Queen Elizabeth Hospital, Woodville 5011, South Australia. Vol. 53, No.6, June 1990
normally cycling women, Halme et al. 2 have demonstrated an increased response to hMG in women with lower body weight so that the day of oocyte collection was earlier than in normal weight subjects with an adverse effect on the frequency of oocyte .fertilization. This retrospective study has analyzed the response to a standardized treatment regime for controlled ovarian hyperstimulation associated with IVF and GIFT with particular reference to the variation in the body weight of subjects. In some contrast to the above studies, the results failed to confirm a major influence of body weight parameters on the ovarian response in these reproductively normal women. MATERIALS AND METHODS
Three hundred sixty-eight subjects referred to the Reproductive Medicine Unit at The Queen Elizabeth Hospital for IVF and GIFT between 1985 and 1988 were included in this retrospective study. All of the selected women had regular reproductive cycles and underwent a standard regime of controlled ovarian hyperstimulation. Patients with severe endometriosis or who had a single ovary which might have precluded a normal ovarian response were excluded, and only cycles of women with a currently known height and weight having their initial cycle of treatment were included. The standard regime of ovarian hyperstimulation used for all subjects in this study utilized CC 100 mg days 5 to 9 and hMG 150 IU (Pergonal, Lewis et al.
Communications-in-brief
1097
Table 1
Relationship Between BMI and Ovarian Response a No. of subjects
Days of treatment
Maximum
BMiindex
27.6
34 114 72 112 36
12.2 ± 1.0 12.6 ± 1.0 12.8± 1.0 13.0 ± 1.0 13.1 ± 1.0
8.9 ± 4.1 9.3 ±4.5 9.7±4.8 8.0±4.2 7.1 ± 3.5
E2
Oocytes recovered
Oocytes fertilized
6.4 ± 3.2 6 6.6±4.0 7.0±4.3 6.4 ± 3.9 4.8 ± 2.6 6
4.7 ± 2.3 5.1 ± 3.3 5.6± 4.4 4.7±.3.8 3.7 ± 2.4
nM/L
a
Values are means± 2 SD.
Serono or Humegon, Organon) days 6, 8, and 10. The ovarian response was carefully monitored on a day-to-day basis, commencing on cycle day 10 by the measurement of plasma estradiol (E 2 ), progesterone, LH, and by abdominal (to 1986) and vaginal ultrasonography of the ovarian follicle(s). Additional daily injections ofhMG (150 IU) were continued until the leading follicles were ::::;15 mm (mean diameter). For the analysis, the number of ampules of hMG required, the peak serum E 2 , the day of hCG or LH surge, and the number of oocytes collected, inseminated, and fertilized were all recorded together with the number of pregnancies achieved. Body mass index (BMI) values for individual women were calculated by the ratio of body weight (kg) divided by the height2 (m 2 ). Total body weight for the group of subjects ranged between 41 and 106 kg (mean ± SD, 61.4 ± 11.5). Height ranged between 148 and 182 em (163 ± 8.2). Regression analysis was not used because the samples had a skewed distribution even after Log transformation. For the purpose of comparison, the sample population was split into five groups according to BMI: underweight ( 90th percentile). Statistical analysis was performed by comparing the median of the groups using MannWhitney nonparametric methods.
6
Significantly different (P < 0.05).
Whereas a trend to increasing E 2 maximum values was identified with both increased number ofhMG ampules used and the increased days of treatment, no significant relationships were found. The data were therefore treated as being homogenous. The relationship between BMI and the ovarian response to hyperstimulation as determined by the serum E 2 level before the day of hCG is shown in Table 1. Whereas the results indicated a tendency for the number of days of treatment required before hCG to increase in subjects with an increased BMI, the difference between the two extreme groups was 27.6 group, the difference was not significant. The mean number of oocytes recovered was greatest in the middle of the BMI range (7.0). The mean number in the BMI < 19.1 group (6.4) was higher (P < 0.05) than in the BMI > 27.6 group (4.8). No differences were apparent in the mean number of oocytes fertilized between groups. Analysis of the data using total body weight in place of BMI demonstrated no significant differences in any parameter examined. A total of 96 pregnancies were established from the IVF and GIFT procedures. The rate of occurrence of clinical pregnancy in the five groups of subjects was 30%, 21%, 24%, 26%, and 20%, respectively. No statistically significant differences between groups were evident.
RESULTS
Given the variation in the hMG dosage used per cycle (range 450 to 1,200 IU) and/or the varying number of days of treatment (range 3 to 9), the data were analyzed with respect to these factors. For the majority of subjects, hCG was given on day 12 (n = 121), day 13 (n = 125), or day 14 (n = 61). 1098
Lewis et al.
Communications-in-brief
DISCUSSION
Weight and body mass are known to be of some importance to the maintenance of regular reproductive cycles. Certainly, either excess weight or marked weight loss can be associated with cycle disturbance. A direct influence of body weight on Fertility and Sterility
ovarian response to exogenous gonadotropins has been shown for subjects requiring induction of ovulation, with obese women requiring larger doses of hMG 3 or CC 4 to effect an optimal response. Data relating to the use of CC and hMG for ovarian hyperstimulation in reproductively normal women are sparse, but Halme et al. 2 found theresponse to hMG was related to body weight, noting particularly the rapid and the hyper-response of underweight women, but with associated poorer oocyte fertilization rates. These authors showed a proven benefit of a reduced dosage schedule over an extended time frame. 5 The number of days of treatment required to reach satisfactory pre-hCG E 2 levels in this study were similar in underweight, normal weight, and obese women. The results thus failed to confirm the earlier findings, although the stimulation regime used differed in that our subjects used CC from days 5 to 9 with hMG supplementation on days 6, 8, and 10, in contrast to the regime of Halme et al., 2 which utilized 150 IU hMG from day 3. It may be, therefore, that the time of commencement of the stimulation regime and perhaps the choice of drugs are critical factors to consider with respect to the effect of body weight or BMI on ovarian responsiveness. It should be also recognized that since a combination of CC and hMG was utilized, the assessment as to whether either of the agents used are weight dependent is difficult. The results of this study may not therefore be applicable to all ovarian hyperstimulation regimes. Whereas the numbers of oocytes recovered were somewhat greater in the underweight than the obese women, these were not different from normal weight subjects. Fertilization rates were comparable in all five groups. When body weight was considered in contrast to BMI, no statistical significance could be derived between any of the parameters measured. It is possible that treating women with normal reproductive function with CC and supraphysiological doses of hMG may well override subtle differences in the levels of hypothalamic pituitary
Vol. 53, No.6, June 1990
ovarian sensitivity. In addition, the large doses used may overcome variability in intramuscular absorption. Drug absorption and metabolic clearance studies related to body weight would be helpful to clarify these possibilities. SUMMARY
A retrospective analysis was performed of 368 normally cycling women treated with a single cycle of a standard ovarian hyperstimulation regime (CC 100 mg days 5 to 9 and hMG 150 IU days 6, 8, and 10) associated with either an IVF or GIFT program. Neither the peak serum E 2 level attained nor the number of days of stimulation required bore a relationship to the BMI or the total body weight of these women. Whereas the mean number of oocytes aspirated from women with BMI < 19.1 was higher (6.4 ± 3.2) compared with obese women (BMI > 27.6, 4.8 ± 2.6), the rate of fertilization was not different for both BMI extremes. It is concluded that factors other than BMI or total body weight have more important influences on the response to hyperstimulation in normal women. Ackrwwledgments. To Barbara Godfrey, M.A.(Oxon), for maintenance of the data base and to the clinical and nursing staff of the Reproductive Medicine Programs for clinical management of the subjects. REFERENCES 1. Reid RL, Van Vugt DA: Weight-related changes in repro-
ductive function. Fertil Steril48:905, 1987 2. Halme J, Hammond MG, Talbert LM, O'Rand M, Bailey L, Sloan C: Positive correlation between body weight, length of human menopausal gonadotropin stimulation, and oocyte fertilization rate. Fertil Steril45:372, 1986 3. Chong AP, Rafael RW, Forte CC: Influence of weight in the induction of ovulation with human menopausal gonadotropin and human chorionic gonadotropin. Fertil Steril46:599, 1986 4. Shepard MK, Balmaceda JP, Leija CG: Relationship of weight to successful induction of ovulation with clomiphene citrate. Fertil Steril32:641, 1979 5. Halme J, Hammond MG, Bailey L, Talbert LM: Lower doses of human menopausal gonadotropin are associated with improved success with in vitro fertilization in women with low body weight. Am J Obstet Gynecol158:64, 1988
Lewis et al.
Communications-in-brief
1099
