(A) Schematic illustrating the layout of the enclosed workstations showing the workflow from the entry of oocytes into the system until the embryos are replaced in the uterus. The incubators act as pass-through hatches linking adjacent workstations. (B) Schematic diagram showing the layout of 2 enclosed workstations connected by the double-doored incubation chambers. (C) Image showing the main features of an enclosed workstation with a built-in stereomicroscope and incubators. There are separate hatches to allow introduction of consumables and removal of waste. (D) Image shows the control panel for setting the temperature of the incubator, workstation and hotplate and the %CO₂ in the incubator and workstation. An alarm sounds when the environmental parameters deviate from the set-points.

(A) A pre-equilibrated organ culture dish was removed from the incubator and placed on the viewing area of either the enclosed workstation (purple) or Class II hood (blue). Time point zero indicates the first measurement taken immediately after placing the dish on the viewing area. The temperature was then measured at 3-minute intervals. The temperature in the Class II cabinet dropped from 36°C±0.1 to 35.2°C±0.3 and then stabilised at 34.9–35.0°C±0.2. In the enclosed workstation, the starting temperature dropped from 37.0–36.7°C±0.1 and stabilised at 36.5–36.6°C±0.1. The lower starting temperature in the Class II cabinet is most likely due to heat loss during transfer from the incubator. In each case 3 independent experiments were conducted. (B) A pre-equilibrated organ culture dish was removed from the incubator and placed on the viewing area of either the enclosed workstation (purple) or Class II cabinet (blue). Time point zero indicates the first measurement taken immediately after placing the dish on the viewing area. There was a gradual increase in pH of the media in the Class II cabinet over the time course of the experiment. By contrast the pH of the media in the enclosed workstation remained in the range 7.30–7.32. In each case 3 independent experiments were conducted. (C) After completing the measurements for Figure 2B, the dish was returned to the incubator and pH measurements taken at 5 minute intervals for 90 minutes. The pH failed to return to 7.31 after 90 minutes in the incubator.

(A) Images show examples of early, expanded and hatched human blastocysts indicating the trophectoderm and the ICM. (B) Development of embryos to the blastocyst stage was compared between the open and enclosed systems. A significantly higher proportion of embryos developed to the blastocyst stage in the enclosed system (P<0.05). (C) Three point moving average of development to the blastocyst stage in consecutive cohorts of twenty embryos. The increase in the proportion of embryos developing to the blastocyst stage coincides with the switch from the open to the enclosed workstations. The two dark bars represent the intermediate period when the cohorts of 20 consisted of embryos cultured in the open and enclosed systems. (D) The proportions of blastocysts graded as early (black), expanded (grey) and hatched (white) on day 6. A significantly higher proportion of blastocysts had undergone hatching by day 6 after culture in the enclosed system compared to the open system. There was a corresponding reduction in the proportion of early blastocysts in the enclosed system (P<0.05). (E) Nuclear counts were performed to compare the total number of cells contained in blastocysts cultured to day 6 or 7 in the open system or enclosed system. The cell count of embryos cultured in the enclosed system was significantly higher on day 6 (128.3±38.6 vs 92.1±43.7; P<0.05) and day 7 (167.7±69.7 vs 106.7±48.8; P<0.01) compared with those cultured in the open system. See Figure S1A–C for an example of nuclear staining in a blastocyst and comparison of nuclear counts between different operators and imaging systems.

(A) Development of human embryos to the blastocyst stage was significantly lower in the open system compared with the enclosed system when incubated in 21% O₂; P<0.05. (B) Nuclear counts were performed to compare the total number of cells contained in human blastocysts cultured to day 6 in the open system or enclosed system in 21% O₂. The cell count of human embryos cultured in the enclosed system in 21% O₂ was significantly higher compared with those cultured in the open system (P<0.05). (C) A significantly higher proportion of mouse embryos developed from the 1 cell to the blastocyst stage in the enclosed system compared with the open system (P<0.05). Both sets of embryos were cultured in 21% O₂. (D) Nuclear counts were performed to compare the total number of cells contained in mouse blastocysts cultured in the open system or enclosed system in 21% O₂. The cell count of mouse embryos cultured in the enclosed system in 21% O₂ was not significantly different compared with those cultured in the open system.

Comparison of the effect of the enclosed system on clinical outcome for consecutive groups of women aged ≤37 years with >10 ovarian follicles undergoing their first cycle of either IVF or ICSI treatment. Embryos from Groups 1 and 2 patients were cultured in the conventional open-fronted system but in different laboratories. Group 3 were conducted in the enclosed system. Letters indicate statistically significant difference. (A) The proportion of transferred embryos resulting in successful implantation, as determined by the presence of a fetal heartbeat at 7 weeks, was significantly higher in Group 3 (31.8%) compared with Group 1 (19.8%) or Group 2 (21.6%); P<0.05. (B) The clinical pregnancy rate per oocyte retrieval was 45.3% for Group 3 compared with 32.2% for Group 1 and 35.6 for Group 2; P<0.05. (C) Comparison of implantation rates in IVF and ICSI treatments in the open and enclosed system.

Cumulative sum of residuals between observed clinical pregnancy rate and that fitted by an overall logistic regression model, plotted against time. During periods of stable performance the plot should have a constant slope: positive when performance is consistently better than average, negative for the opposite. The vertical lines mark the group boundaries. Regions with positive slopes indicate performance consistently better than average, regions with negative slopes indicate the opposite. The start of the third phase, group 3, coincides precisely with a change from a negative to positive slope and a statistically significant improvement in performance (p<0.001). There is no change in performance between the first and second phases or within any phase.