Prognostic value of the postcoital test for spontaneous pregnancy

Transcription

1 PREGNANCY Prognostic value of the postcoital test for spontaneous pregnancy Esther Leushuis, M.D., M.Sc., a,b Jan Willem van der Steeg, M.D., Ph.D., b Pieternel Steures, M.D., Ph.D., b Carolien Koks, M.D., Ph.D., c Jur Oosterhuis, M.D., Ph.D., d Petra Bourdrez, M.D., e Patrick M. M. Bossuyt, Ph.D., f Fulco van der Veen, M.D., Ph.D., b Ben Willem J. Mol, M.D., Ph.D., b,f and Peter G. A. Hompes, M.D., Ph.D., a on behalf of the CECERM study group a Department of Obstetrics and Gynecology, Vrije Universiteit Medical Center, Amsterdam; b Center for Reproductive Medicine, Academic Medical Center, Amsterdam; c Department of Obstetrics and Gynecology, Maxima Medical Center, Veldhoven; d Department of Obstetrics and Gynecology, Medisch Spectrum Twente, Enschede; e Department of Obstetrics and Gynecology, Vie Curie Medical Center, Venlo; and f Department of Clinical Epidemiology and Biostatistics, Academic Medical Center, Amsterdam, the Netherlands Objective: To evaluate the capacity of the postcoital test (PCT) to predict spontaneous pregnancy in a large cohort study of subfertile couples. Design: Prospective study. Setting: Department of reproductive medicine of 38 hospitals in the Netherlands. Patient(s): Between January 2002 and February 2004, we prospectively included consecutive subfertile couples who had not been evaluated previously for subfertility. Intervention(s): We estimated the contribution of the PCT result to the existing prediction model for spontaneous pregnancy by calculating the adjusted hazard ratio (HR) of an abnormal PCT result. We constructed a second prediction model (PCT model) based on the reference model including the PCT. Main Outcome Measure(s): Primary endpoint in this study was ongoing pregnancy. We evaluated the performance of the PCT model in comparison with the reference model by calculating goodness of fit, discrimination, calibration, and the net reclassification improvement. Result(s): We included 3,021 couples of whom 537 (18%) had a spontaneous pregnancy and 55 (1.8%) a nonsuccessful pregnancy; 1,316 (44%) started treatment within 12 months, 824 (27%) neither started treatment nor became pregnant, and 289 (10%) became lost to follow-up within 12 months. The adjusted HR for an abnormal PCTwas 0.76 (95% confidence interval [CI]: 0.62 to 0.94). The adjusted HR for an abnormal PCTwas 0.63 (95% CI: 0.47 to 0.84) in case of no spermatozoa, 0.81 (95% CI: 0.57 to 1.2) in case of nonmotile spermatozoa, and 1.2 (95% CI: 0.8 to 1.8) in case of motile, nonprogressive spermatozoa. Adding the PCT result to the reference model did not improve goodness of fit. Discrimination was equally poor for the PCT model and the reference model. The calibration plots of both models showed comparably good calibration. The net reclassification improvement of the predictions of the PCT model compared with the reference model was 1.1%. Conclusion(s): This study demonstrated that the PCT has prognostic value but does not add substantially to a prognostic model for spontaneous pregnancy. (Fertil Steril Ò 2011;95: Ó2011 by American Society for Reproductive Medicine.) Key Words: Postcoital test, prognosis, prediction model, hazard ratio, net reclassification improvement Received August 30, 2010; revised February 14, 2011; accepted February 17, 2011; published online March 27, E.L. has nothing to disclose. J.W.v.d.S. has nothing to disclose. P.S. has nothing to disclose. C.K. has nothing to disclose. J.O. has nothing to disclose. P.B. has nothing to disclose. P.M.M.B. has nothing to disclose. F.v.d.V. has nothing to disclose. B.W.J.M. has nothing to disclose. P.G.A.H. has nothing to disclose. CECERM study group (Collaborative Effort for Clinical Evaluation in Reproductive Medicine) Investigators, participating centers in the Netherlands: Y. M. van Kasteren, Alkmaar, Medisch Centrum Alkmaar; P. F. M. van der Heijden, Almelo, Twenteborg Ziekenhuis; W. A. Sch ols, Amersfoort, Meander Medisch Centrum; M. H. Mochtar, Amsterdam, AcademischMedischCentrum;G.L.M.Lips,Amsterdam,BovenIJZiekenhuis;J. Dawson, Amsterdam, St. Lucas Adreas Ziekenhuis; H. R. Verhoeve, Amsterdam, Onze Lieve Vrouwe Gasthuis; S. Milosavljevic, Amsterdam, Slotervaart Ziekenhuis; P. G. A. Hompes, Amsterdam, Vrij Universiteit Medisch Centrum; L. J. van Dam, Apeldoorn, Gelre Ziekenhuis; A. V. Sluijmer,Assen,WilhelminaZiekenhuis;H.E.Bobeck,Beverwijk,Rode Kruis Ziekenhuis; R. E. Bernardus, Blaricum, Ziekenhuis Gooi-Noord; M. C. S. Vermeer, Breda, Amphia Ziekenhuis Breda; J. P. D orr, Den Haag, Medisch Centrum Haaglanden, Locatie Westeinde; P. J. Q. van der Linden, Deventer, Ziekenhuis Deventer; H. J. M. Roelofs, Dordrecht, A. Schweitzer Ziekenhuis; J. M. Burggraaff, Emmen, Scheper Ziekenhuis; G. J. E. Oosterhuis, Enschede, Medisch Spectrum Twente; M. H. Schouwink, Geldrop, St. Anna Ziekenhuis; P. X. J. M. Bouckaert, Heerlen, Atrium Medisch Centrum; F. M. C. Delemarre, Helmond, Elkerliek Ziekenhuis; C. J. C. M. Hamilton, s Hertogenbosch, Jeroen Bosch Ziekenhuis; M. van Hoven, Hilversum, Ziekenhuis Hilversum; M. H. Emanuel, Hoofddorp, Spaarne Ziekenhuis; C. N. M. Renckens, Hoorn, Westfries Gasthuis; J. A. Land, Maastricht, Academisch Ziekenhuis Maastricht; J. H. Schagen- Van Leeuwen, Nieuwegein, St. Antonius Ziekenhuis; J. A. M. Kremer, Nijmegen, UMC St. Radboud; C. van Katwijk, Purmerend, Waterland Ziekenhuis; M. H. A. van Hooff, Rotterdam, St. Franciscus Gasthuis; H. J. H. M. Van Dessel, Tilburg/Waalwijk, TweeSteden Ziekenhuis; F. J. M. Broekmans, Utrecht, UMC Utrecht; H. J. L. A. Ruis, Veghel, Ziekenhuis Bernhoven; C. A. M. Koks, Veldhoven, Maxima Medisch Centrum; P. Bourdrez, Venlo/Venray, Vie Curi Medisch Centrum; W. W. J. Riedijk, Zaandam, Zaans Medisch Centrum; B. J. Cohlen, Zwolle, Isala Klinieken. Reprint requests: Esther Leushuis, M.Sc., M.D., Academic Medical Center, Center of Reproductive Medicine, Room H4-238, Department of Obstetrics and Gynecology, Meibergdreef 9, 1105 AZ Amsterdam, the Netherlands ( e_leushuis@hotmail.com) Fertility and Sterility â Vol. 95, No. 6, May /$36.00 Copyright ª2011 American Society for Reproductive Medicine, Published by Elsevier Inc. doi: /j.fertnstert

2 Since the original description of the postcoital test (PCT) in 1866 by J. Marion Sims (1) and the reintroduction by M. Huhner in 1913 (2) its use has become widespread in the evaluation of subfertile couples. Many authorities considered the PCT as the cornerstone of the fertility evaluation (3). Although the PCT is used commonly, the use of this test in the basic fertility workup has been subject to debate over the last 10 years because of conflicting data (4 6). On the one hand, the PCT enables identification of couples with a cervical factor and avoids misclassifying these couples as having unexplained infertility, which has important implications for treatment. An abnormal PCT result decreases the probability of treatment-independent pregnancy twofold to threefold (5, 7 9). On the other hand, routine use of the PCT would only lead to more interventions without an increase in pregnancy rates (10). Recently, we validated a prediction model for spontaneous pregnancy in a large cohort of subfertile couples (11). In this validation, the prediction model including female age, duration of subfertility, primary subfertility, sperm motility, and referral status had good predictive performance, but adding PCT results to this model did not improve its performance. Thus, at present, data on the prognostic value of the PCT for spontaneous pregnancy in subfertile couples are inconclusive. The aim of this study was to evaluate the prognostic value of the PCT in a large prospective multicenter cohort of subfertile couples in relation to an existing, well-validated prediction model for spontaneous pregnancy. MATERIALS AND METHODS Validation Cohort The study was designed as a prospective cohort study performed in 38 hospitals in the Netherlands. The Institutional Review Board of the Academic Medical Center approved the study, and local approval was obtained from the Board of Directors from each participating hospital. Between January 2002 and February 2004, we included consecutive subfertile couples who had not been evaluated previously for subfertility (11). In couples who had an unfulfilled wish for a child and had 1 year with regular unprotected intercourse and in whom the woman had a regular cycle, a basic fertility workup was performed, consisting of a fertility history, semen analysis, PCT, assessment of ovulation, and assessment of the Fallopian tubes according to the guidelines of the Dutch Society of Obstetrics and Gynecology (2004). The performance of the PCT was standardized as described in detail in the study protocol for all participating centers. At least one PCT was performed during the infertility workup (5, 7, 9, 11). The PCT could be planned on the basis of the basal body temperature and cycle length or on repetitive ultrasound findings. The PCT was judged to be normal if at least one progressively motile spermatozoon was seen in one of five high-power fields at 400 magnification. The PCTwas judged abnormal if motile, nonprogressive spermatozoa or nonmotile or no spermatozoa were seen in one of the five high-power fields at 400 magnification. Follow-up The follow-up of the couples is described in full detail in the publication of van der Steeg et al. (11). Primary endpoint was spontaneous conception resulting in an ongoing pregnancy. Spontaneous ongoing pregnancy was defined as the presence of fetal cardiac activity at transvaginal sonography at a gestational age of at least 12 weeks, resulting from a treatmentindependent conception. Time to pregnancy was considered censored at the moment treatment had been started or at the last date of contact during follow-up, when the couple had no ongoing pregnancy. For all couples lost to follow-up, the general practitioner was sent a questionnaire on the most recent fertility status of the couple. Data Analysis Missing data for the predictive variables were imputed ( filled in ), because deleting them would lead to a loss of statistical power and potentially biased results in multivariable analysis (12, 13). The technique of imputation and details of this specific dataset are described in a previous publication (11). The Hunault model for predicting spontaneous pregnancy was considered the reference model (14). This model includes five prognostic variables: female age, duration of subfertility, female subfertility being primary or secondary, percentage motile spermatozoa of the first semen analysis, and referral status. Referral by a gynecologist was established as a confounding factor in the original publication. We then evaluated the additional value of the PCT by building a new multivariable Cox model: the PCT model. This PCT model includes the variables from the reference model and, additionally, the PCT outcome. We did not recalculate the weights for the variables in the reference model, as this would lead to a loss in power and an increased risk of capitalization on chance: an artificial increase in predictive power that is due only to chance fluctuations in the variables. We therefore used the linear combination of all variables in the reference model. In other words, for each couple we calculated the sum of the variables in the reference model, each variable weighted by the respective coefficient as estimated by Hunault et al. (14). We then estimated two coefficients for our PCT model using multivariable Cox regression analysis: one coefficient for the sum of the weighted variables of the reference model and a second coefficient for the PCT outcome. We transformed the coefficients into hazard ratios (HR) to facilitate interpretation. We performed a separate analysis to assess the prognostic value of each separate abnormal PCT category relative to a normal PCT: motile, nonprogressive spermatozoa and nonmotile or no spermatozoa. Evaluation of Performance We compared the goodness of fit of the PCT model to our data with the goodness of fit of the reference model using the generalized likelihood ratio test. This test evaluates whether adding the PCT to the reference model leads to a significant improvement in goodness of fit. We then used three concepts to compare the performance of the PCT model with that of the reference model. First, we evaluated the discriminative power of the models. Discrimination refers to the ability to distinguish couples who will conceive from those who will not (15). Second, we evaluated the calibration of the models. Calibration is the level of correspondence between the calculated pregnancy probabilities and the observed proportion of pregnancies. For a comprehensive explanation of discrimination and calibration and the interpretation of performance we reference our recent publications (11, 15). Third, we determined the degree of reclassification between the reference model and the PCT model by calculating the net reclassification improvement (16, 17). This method is based on the difference between two models in the individual calculated probabilities of a pregnancy and the occurrence of a pregnancy in couples with probabilities that are discordant between the two models. Calculating the net reclassification improvement requires a priori meaningful risk categories. We used 0% to 30%, 30% to 40%, and >40% for the probability of spontaneous pregnancy (11). In net reclassification improvement, only those changes in estimated probabilities that imply a change from one category to another (reclassification) are considered. The reclassification of couples who develop and who do not develop a pregnancy should be considered separately. Any upward movement in categories for couples who achieved a pregnancy implies improved classification, and any downward movement indicates worse reclassification. The interpretation was opposite for couples who did not achieve a pregnancy. The net improvement in reclassification was quantified as a sum of differences in proportions of couples moving up minus the proportion moving down for couples who develop events, Fertility and Sterility â 2051

3 TABLE 1 Baseline characteristics of the 3,021 included subfertile couples. Characteristic Mean or median 5th 95th percentile Female age (y) Male age (y) Duration of subfertility (y) (median) Subfertility, primary (n) 2,013 67% Sperm motility (grade A WHO %) PCT, a normal (n) 2,050 68% Note: WHO ¼ World Health Organization. a Postcoital test. and the proportion of couples moving down minus the proportion moving up for couples who do not develop events. RESULTS We completed the fertility workup of 7,860 subfertile couples. Of these, 948 had a severe male factor, 311 had two-sided tubal pathology, 1,311 had one-sided tubal pathology, and 2,642 had other reasons for exclusion (i.e., prior fertility treatment), leaving 3,021 couples for inclusion. The baseline characteristics of these couples are presented in Table 1. We were able to complete follow-up for 2,741 couples (90%). Of the 3,021 included couples, 537 (18%) had a spontaneous ongoing pregnancy within 1 year, including 10 multiple pregnancies (0.3%). Within 12 months 1,316 (44%) of all couples had started treatment, whereas 824 (27%) neither had started treatment yet nor had become pregnant within the follow-up period. The followup status of all patients at 12 months is shown in Figure 1. The mean probability of a spontaneous pregnancy as calculated with the reference Hunault model without PCT was 0.32 (5th and 95th percentiles: 0.16 and 0.52). For the PCT model, the mean calculated probability was 0.29 (5th and 95th percentiles: 0.16 and 0.45). The adjusted HR of an abnormal PCT result in the PCT model was 0.76 (95% confidence interval [CI]: 0.62 to 0.94). The adjusted HR of an abnormal PCT was 0.63 (95% CI: 0.47 to 0.84) in case of no spermatozoa, 0.81 (95% CI: 0.57 to 1.2) in case of nonmotile spermatozoa, and 1.2 (95% CI: 0.8 to 1.8) in case of motile, nonprogressive spermatozoa, each relative to a normal PCT. Evaluation of Performance Goodness of fit was not improved significantly when the PCT model was compared with the reference model (difference in 2 log likelihood: 7, P>.05 [degrees of freedom 3]). Discrimination was FIGURE 1 Follow-up status of all patients at 12 months. Fertility workup completed 7,860 Exclusion (causal pathology): - Ovulation disorder 938 (12%) - Severe male factor (VCM<3*10 6 ) - Double-sided tubal pathology 948 (12%) 311 ( 4%) Exclusion (other): - Referred by a gynecologist - Prior fertility treatment 680 (8.7%) 627 (8.0%) - One-sided tubal pathology 1,335 (17%) External validation data 3,021 Follow-up Spontaneous pregnancy within 1 year 592 ( 20%) - Ongoing pregnancy, singleton 527 ( 17%) - Ongoing pregnancy, multiple 10 (0.3%) - Miscarriage 51 (1.7%) - Ectopic pregnancy 4 (0.1%) No spontaneous pregnancy within 1 year 2,429 (80%) - Treatment within 1 year 1,316 (44%) - No spontaneous pregnancy within 1 year 824 (27%) - Lost to follow-up 289 (10%) 2052 Leushuis et al. Prognostic value of the postcoital test Vol. 95, No. 6, May 2011

4 comparable for the PCT model (area under the curve: 0.64; 95% CI: 0.61 to 0.66) and the reference model (area under the curve: 0.63; 95% CI: 0.60 to 0.65), a nonsignificant difference. The calibration plots of both models showed good calibration (Fig. 2). The calibration of the PCT model was slightly better than the calibration of the reference model. The addition of the PCT to the reference model reduced the correct classification of couples who achieved a pregnancy with 19.0% but improved the correct classification of the couples who did not achieve a pregnancy with 17.9%. The net reclassification improvement of the predictions of the PCT model compared with the reference model was 1.1%. The PCT model showed a minor net reduction in the correct classification of couples who did and did not achieve a pregnancy compared with the reference model. The details of the classification per model per couple are shown in Table 2. FIGURE 2 (A) Calibration plot of the reference model. (B) Calibration plot of the PCT model. 1 0,9 0,8 Observed proportion 0,7 0,6 0,5 0,4 0,3 0,2 0, ,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 Predicted probability 1 0,9 0,8 0,7 Observed proportion 0,6 0,5 0,4 0,3 0,2 0, ,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 Predicted probability Fertility and Sterility â 2053

5 DISCUSSION This study showed that subfertile couples with an abnormal PCT result have a lower probability of a spontaneous pregnancy, especially if no sperm are found in the cervical mucus. Despite the prognostic information of the PCT, addition of the PCT outcome as a prognostic variable to a validated prediction model for spontaneous pregnancy did not improve the fit of this model significantly and even led to a nonsignificant reduction in the correct classification of couples who did and did not achieve a pregnancy. The prognostic value of the PCT for spontaneous pregnancy has so far been assessed in multivariable models and in one randomized controlled trial (5, 7, 9, 11, 14, 18, 19). None of these studies was designed specifically to assess the prognostic value of the PCT to the already established prognostic factors for spontaneous pregnancy. The first study of 726 subfertile couples presented an adjusted HR for an abnormal PCT of 0.26 (95% CI 0.17 to 0.40) for the prediction of live birth, but two out of four variables selected for the final model tubal defects and anovulation should have been excluded because of the manifestly reduced chances of spontaneous conception in couples with these factors (7). Moreover, the HR of the PCT in this study could have been overestimated because of insufficient adjustment for the established prognostic variables. A second study, in 996 subfertile couples, reported that a progressive PCT was the most important prognostic variable of all, with a HR of 0.23 (95% CI: 0.14 to 0.4) (5), but the model derived from this study demonstrated poor calibration at external validation (11). The model underestimated the pregnancy rates in couples with a poor prognosis (0% to 20%) and overestimated them in couples with a good prognosis (R40%). A third study was a randomized clinical trial. The authors of this trial concluded that the routine use of the PCT leads to more interventions without an associated increase in pregnancy rates (19). This study was criticized because a substantial number of participating women were anovulatory and because no specific recommendations were made on how to manage a couple with an abnormal PCT (20 23). A fourth study in 207 couples evaluated the prognostic value of the PCT in a multivariable model, including the variables duration of subfertility, age, PCT result, and motile sperm concentration (9). Surprisingly, this study showed no prognostic value for woman s age or motile sperm concentration. The added prognostic value of the PCT, relative to the established factors for spontaneous pregnancy, could not be assessed, possibly because of the size of the study population but primarily because of a lack of complete data to enable assessment of all prognostic factors and model performance. The last study to evaluate the PCT outcome together with the established factors generated two models for the prediction of spontaneous pregnancy, one with and one without the PCT, showing inferior performance for the model without the PCT. However, the derivation of the model with the PCT was based on a smaller cohort of patients (n ¼ 1,398) than the model without the PCT (n ¼ 2,459), which could have led to a less accurate estimation of regression coefficients (14). Further, these models were validated in a prospective validation cohort study. The performance of the model with the PCT again was inferior to the model with the PCT, but the primary aim of that study was to evaluate the performance of the model without the PCT, and the techniques used in the external validation for the assessment of performance were limited (11). One of the strengths of our study is that we were able to assess the specific value of the PCT in addition to the already established factors for the prediction of spontaneous pregnancy in a large prospective multicenter cohort of subfertile couples. We excluded couples with known causes of subfertility that manifestly reduce chances of spontaneous conception. We were able to include a large TABLE 2 Net reclassification improvement of the PCT model compared with the reference model. PCT model, frequency (row percent) <30%, n (%) 30% to 40% >40% Reference model No. % No. % No. % Total (no.) Couples who achieved a pregnancy <30% a a % to 40% 41 b a >40% 0 b 0 61 b Total Couples who did not achieve a pregnancy <30% 1, b 0 0 b 0 1,113 30% to 40% 228 a b >40% 0 a a Total 1, ,217 Note: Net reclassification for couples who achieved a pregnancy is a deterioration of 19% and for the couples who did not achieve a pregnancy is an improvement of 17.9%. This results in an overall net reclassification improvement of the PCT model of 1.1%. a Deterioration of classification for the PCT model. b Improvement of classification for the PCT model Leushuis et al. Prognostic value of the postcoital test Vol. 95, No. 6, May 2011

6 set of variables and used the firm evidence gained from an established prediction model to build onto. The statistical analysis of this study allowed for the time of spontaneous chance to conception by censoring the outcomes, even if the couples had undergone treatment within 1 year. In an ideal world it would have been better to allow all couples the opportunity of spontaneous conception over the same time period; nowadays this is difficult to achieve. The outcome of couples that were lost to follow-up can be regarded as being missing completely at random. There is no reason to assume that positive results are systematically more often missing than negative results. The couples that were lost to follow-up were classified as not becoming pregnant. The PCT allows identification of couples where the spermatozoa do not penetrate the cervical mucus. In a previous randomized clinical trial, we showed that, in couples with an abnormal PCT but a good prognosis, IUI in the unstimulated cycles was superior over no treatment (24). Similarly, in couples with an abnormal PCT but otherwise also a poor prognosis, IUI without controlled ovarian hyperstimulation (COH) leads to pregnancy rates that are comparable with those obtained by IUI with COH. IUI without COH should therefore be the treatment of first choice in these couples, as it reduces costs and the risk of multiple pregnancies, without compromising pregnancy rates (25). Although we found that the PCT does not help in establishing a prognosis for spontaneous conception, it might identify couples who do not conceive as a consequence of sexual dysfunction, who may benefit from sexual therapy or insemination in the natural cycle. If we abandon the PCT, we should realize that we potentially do not identify some couples with subfertility who actually have sexual dysfunction. The PCT is the only test in the fertility workup that endeavors to assess sexual function in vivo, thus allowing identification of subfertile couples who do not conceive because they fail to have successful intercourse, whereas their biologic components necessary for conception semen, ovulation, and tubal function are all normal. Studies reporting on the PCT showed that in 7% to 16% no spermatozoa were seen (26 28). In these couples no pregnancies were observed. No spermatozoa were seen in 17% of all PCTs in our study, but 49 pregnancies were seen. The PCT or a variant might be part of the fertility workup in a research setting. In summary, our study demonstrated that the postcoital test has prognostic value but does not significantly add value to a prognostic model for spontaneous pregnancy. This adds to a previous randomized clinical trial that indicated no benefit of the PCT. The test adds no prognostic information to existing models, and the predictive information from the PCT is already incorporated in other variables based on semen analysis. REFERENCES 1. Sims JM. Clinical notes on uterine surgery (with special reference to the management of the sterile condition). London: Robert Hardwicke; Hunher M. Sterility in the male and female and its treatment. New York: Rebman; Glatstein IZ, Best CL, Palumbo A, Sleeper LA, Friedman AJ, Hornstein MD. The reproducibility of the postcoital test: a prospective study. Obstet Gynecol 1995;85: Griffith CS, Grimes DA. The validity of the postcoital test. Am J Obstet Gynecol 1990;162: Eimers JM, te Velde ER, Gerritse R, Vogelzang ET, Looman CW, Habbema JD. The prediction of the chance to conceive in subfertile couples. Fertil Steril 1994;61: Oei SG. The postcoital test: a controversial investigation. Eur J Obstet Gynecol Reprod Biol 1998;77: Snick HK, Snick TS, Evers JL, Collins JA. The spontaneous pregnancy prognosis in untreated subfertile couples: the Walcheren primary care study. Hum Reprod 1997;12: Oei SG, Helmerhorst FM, Bloemenkamp KW, Keirse MJ. Effect of the postcoital test on the sexual relationship of infertile couples: a randomized controlled trial. Fertil Steril 1996;65: Glazener CM, Ford WC, Hull MG. The prognostic power of the post-coital test for natural conception depends on duration of infertility. Hum Reprod 2000;15: Oei SG, Helmerhorst FM, Keirse MJ. Routine postcoital testing is unnecessary. Hum Reprod 2001;16: van der Steeg JW, Steures P, Eijkemans MJ, Habbema JD, Hompes PGA, Broekmans FJ, et al. Pregnancy is predictable: a large-scale prospective external validation of the prediction of spontaneous pregnancy in subfertile couples. Hum Reprod 2007;22: Schafer JL. Analysis of incomplete multivariate data. London: Chapman & Hall; Little RJA, Rublin DB. Statistical analysis with missing data. New York: Wiley; Hunault CC, Habbema JD, Eijkemans MJ, Collins JA, Evers JL, te Velde ER. Two new prediction rules for spontaneous pregnancy leading to live birth among subfertile couples, based on the synthesis of three previous models. Hum Reprod 2004;19: Leushuis E, van der Steeg JW, Steures P, Bossuyt PM, Eijkemans MJ, van der Veen F, et al. Prediction models in reproductive medicine: a critical appraisal. Hum Reprod Update 2009;15: Pencina MJ, D Agostino RB Sr, D Agostino RB Jr, Vasan RS. Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyond. Stat Med 2008;27: Meigs JB, Shrader P, Sullivan LM, McAteer JB, Fox CS, Dupuis J, et al. Genotype score in addition to common risk factors for prediction of type 2 diabetes. N Engl J Med 2008;359: Glazener CM, Kelly NJ, Weir MJ, David JS, Cornes JS, Hull MG. The diagnosis of male infertility prospective time-specific study of conception rates related to seminal analysis and post-coital sperm-mucus penetration and survival in otherwise unexplained infertility. Hum Reprod 1987;2: Oei SG, Helmerhorst FM, Bloemenkamp KW, Hollants FA, Meerpoel DE, Keirse MJ. Effectiveness of the postcoital test: randomised controlled trial. Br Med J 1998;317: Bossuyt PM, Lijmer JG, Mol BW. Randomised comparisons of medical tests: sometimes invalid, not always efficient. Lancet 2000;356: Hull MG, Evers JL. Postcoital testing. Criterion for positive test was not given. Br Med J 1999;318: Cohlen BJ, te Velde ER, Habbema JD. Postcoital testing. Postcoital test should be performed as routine infertility test. Br Med J 1999;318: Hendry WF. Postcoital testing. Male partner should be assessed. Br Med J 1999;318: Steures P, van der Steeg JW, Hompes PGA, Bossuyt PM, Habbema JD, Eijkemans MJC, et al. Effectiveness of intrauterine insemination in subfertile couples with an isolated cervical factor: a randomized clinical trial. Fertil Steril 2007; 88: Steures P, van der Steeg JW, Hompes PGA, Habbema JDF, Eijkemans MJC, Broekmans FJ, et al. Intrauterine insemination with controlled ovarian hyperstimulation versus expectant management for couples with unexplained subfertility and an intermediate prognosis: a randomised clinical trial. Lancet 2006;368: Hull MG, Savage PE, Bromham DR. Prognostic value of the postcoital test: prospective study based on timespecific conception rates. Br J Obstet Gynaecol 1982;89: Moghissi KS, Sacco AG, Borin K. Immunologic infertility. I. Cervical mucus antibodies and postcoital test. Am J Obstet Gynecol 1980;136: Giner J, Merino G, Luna J, Aznar R. Evaluation of the Sims-Huhner postcoital test in fertile couples. Fertil Steril 1974;25: Fertility and Sterility â 2055