Background Numerous randomised studies have reported pregnancy outcome in women who received acupuncture during their in vitro fertilisation (IVF) treatment cycle.Objective The objective of this study was to conduct a systematic review with meta-analysis of the trials of acupuncture during IVF treatment on the outcomes of clinical pregnancy and live birth rates.Search strategy Searches were conducted in MEDLINE, EMBASE, Cochrane Library, ISI Proceedings and SCISEARCH.Selection criteria All randomised controlled trials that evaluated the effects of acupuncture compared with no treatment or sham acupuncture in women undergoing IVF-intracytoplasmic sperm injection treatment were included.Data collection and analysis Study selection, quality appraisal and data extraction were performed independently and in duplicate. A sensitivity analysis was conducted where the meta-analysis was restricted to trials in which sham acupuncture was used in the control group. Meta-regression analysis was used to explore the association between study characteristics and pregnancy rates.Main results Thirteen relevant trials, including a total of 2500 women randomised to either acupuncture or control group, were identified. No evidence of publication bias was found (Begg's test, P = 0.50). Five trials (n = 877) evaluated IVF outcome when acupuncture was performed around the time of transvaginal oocyte retrieval, while eight trials (n = 1623) reported IVF outcome when acupuncture was performed around the time of embryo transfer (ET). Meta-analysis of the five studies of acupuncture around the time of egg collection did not show a significant difference in clinical pregnancy (relative risks [RR] = 1.06, 95% CI 0.82-1.37, P = 0.65). Meta-analysis of the eight studies of acupuncture around the time of ET showed no difference in the clinical pregnancy rate (RR = 1.23, 95% CI 0.96-1.58, P = 0.1). Live birth data were available from five of the eight studies of acupuncture around the time of ET. Metaanalysis of these studies did not show a significant increase in live birth rate with acupuncture (RR = 1.34, 95% CI 0.85-2.11). Using meta-regression, no significant association between any of the studied covariates and clinical pregnancy rate was found (P > 0.05 for all covariates).Conclusion Currently available literature does not provide sufficient evidence that adjuvant acupuncture improves IVF clinical pregnancy rate.

Using a bilateral medial preoptic area (MPOA) infusion system in conscious female rats we have investigated the role of the GABA system in this area on the proestrous luteinizing hormone (LH) surge. Fifteen-minute blood samples for LH estimation were taken throughout the afternoon of proestrus from female rats exhibiting 4-day oestrous cycles and implanted at least 2 weeks prior with cerebral guide cannulae. Between 15:00 and 17:00 h rats received an infusion (1 µl/30 min) of artificial cerebro-spinal fluid (n = 7), 10 µM GABA (n = 6) or 10 µM bicuculline methiodide (BMI, n = 6). Animals infused with GABA failed to exhibit an LH surge, while BMI-treated animals displayed an LH surge which was not significantly different to controls. These data show that on the afternoon of proestrus, there are no tonic modulatory actions of the GABA system, acting through the GABAA receptor, on neural elements controlling the LH surge in the MPOA. If however, GABA levels are elevated in the MPOA at this time then the LH surge is blocked. In conjunction with data from correlative studies showing a decrease in endogenous GABA release prior to the LH surge, we suggest that this fall in activity is an essential component of the LH surge mechanism.

Blood samples were taken from conscious, chronically-catheterized rats during parturition for measurement of oxytocin by specific radioimmunoassay. After the birth of the 3rd pup, rats were allowed to remain in their nesting cage (undisturbed rats) or were transferred for 45 min to a glass bowl (disturbed rats); at the time of transfer, rats were given an i.v. injection of the opioid antagonist naloxone or saline vehicle. Subsequent parturition was prolonged in saline-treated disturbed rats, but not in naloxone-treated disturbed rats. Parturition was significantly hastened in naloxone-treated undisturbed rats. Naloxone injections were followed by a large rise in plasma oxytocin concentrations in disturbed and undisturbed rats. We conclude, from a statistical analysis of the relationship within experimental groups between plasma oxytocin concentration and speed of parturition, that the effects of disturbance and of naloxone upon parturition may be accounted for, at least in part, by their effects upon oxytocin release. However, the effects of disturbance on parturition may not be mediated entirely by activation of opioid pathways. Naloxone did not potentiate oxytocin release in non-pregnant rats, or on Day 1 post partum, but did potentiate oxytocin release on Day 22 of pregnancy even in rats before the onset of parturition. Endogenous opioid pathways regulating oxytocin release therefore appear to be active during late pregnancy and during parturition itself.

Two experiments were undertaken to investigate the way that fasting impairs LH secretion and to assess whether endogenous opioid mechanisms might be responsible for the impairment. In the first experiment, pulsatile LH secretion was measured in a total of 51 chronically ovariectomized female rats. Initially 29 rats were subjected to food withdrawal for 24, 48, 72 or 120 h before the experiment. When compared with data collected from eight unfasted control rats, the 120-h fast was found to reduce significantly the mean peak and trough values of the LH pulses measured. However, in a subsequent study, the inhibition of pulsatile LH secretion by a 120-h fast was prevented in a group of eight rats given the opioid antagonist naloxone hydrochloride before the start of the blood-sampling period. Naloxone was without effect on pulsatile LH secretion in eight unfasted control rats. In the second experiment, plasma LH concentrations were measured before and after unilateral electrical stimulation of the ventral noradrenergic tract (VNAT) in ovariectomized female rats pretreated with oestradiol benzoate. In 17 rats VNAT stimulation caused a significant rise in plasma LH, but after a 72-h fast this rise was significantly less than in unfasted control rats. However, pretreatment of fasted rats with naloxone (n = 9) significantly enhanced the VNAT-stimulated release of LH to the control values. Naloxone did not potentiate VNAT-stimulated LH release in unfasted animals (n = 6) or LH release in control unstimulated rats (n = 12). The experiments indicate that both pulsatile LH secretion, and LH release caused by VNAT stimulation, are impaired by an acute fast.(ABSTRACT TRUNCATED AT 250 WORDS)

The role of GABA neurones in the medial preoptic area (MPOA) in regulating the activity of the luteinising hormone-releasing hormone (LHRH) neurones projecting to the median eminence was investigated in the conscious ovariectomised rat. Plasma luteinising hormone (LH) concentrations were measured while (1) endogenous GABA release from the MPOA was monitored with the technique of microdialysis, or (2) activity at the GABA receptor was modulated by local infusions into the MPOA. Microdialysis studies revealed a fluctuating level of GABA release in the MPOA which did not correlate with pulsatile LH secretion. Infusion of 10 microM GABA (n = 8) or bicuculline methiodide (BMI, n = 6) into the MPOA, at a rate of 1 microliter/30 min, significantly inhibited mean LH concentrations (P less than 0.05-0.001) and LH pulse frequency (P less than 0.05-0.001) compared with controls (n = 8). LH pulse amplitude was not significantly altered by infusion of GABA (P greater than 0.05) while too few pulses were found after BMI treatment to enable statistical analysis. Infusions of GABA into the ventral half of the MPOA had a more significant inhibitory effect upon LH secretion compared with dorsal infusions (P = 0.012). A similar relationship did not exist for BMI infusions. These results show that acute changes in preoptic GABA receptor occupancy result in disruption of pulsatile LH secretion in the ovariectomised rat. This suggests that GABA neurones provide a tonic input important for the functional integrity of the neural network controlling LH secretion.(ABSTRACT TRUNCATED AT 250 WORDS)