Prior to 1976, most bovine embryos were collected via mid-line laparotomy or, less commonly, via a flank incision. In that year, several groups published efficacious methods for non-surgical (transcervical) recovery of embryos, and the industry changed to these procedures rather abruptly (see Betteridge, 1977, for review).
In most cases, embryos are recovered six to eight days after the beginning of oestrus (day 0). Embryos can be recovered non-surgically as early as four days after oestrus from some cows, but prior to day 6 recovery rates are lower than on days 6 to 8. Embryos can also be recovered on days 9 to 14 after oestrus; however, they hatch from the zona pellucida on day 9 or 10, making them more difficult to identify and isolate and more susceptible to infection. After day 13, embryos elongate dramatically and are sometimes damaged during recovery or become entangled with each other. Procedures for cryopreservation and bisection have been optimized for day 6–8 embryos, which is another reason for choosing this time. A small percentage of embryos remain in the oviduct after day 7. Unfortunately these are not recoverable with current non-surgical procedures.
The first step in non-surgical recovery is to palpate the ovaries per rectum to estimate the number of corpora lutea. This is very difficult to do accurately if there is a large response to superovulation, although it is not critical to determine how large this response is. Even when only two or three corpora lutea are palpated by skilled personnel, occasionally four or five embryos are recovered. However, it is exceedingly rare to obtain embryos if there are no palpable corpora lutea by day 7. Under most circumstances, cows with no response are not worth flushing, although occasionally an embryo is recovered. It is rare to recover more than one embryo from cows with one palpable corpus luteum. In many situations, donors are palpated the day before recovery or the morning of recovery so that logistical plans can be made, for example, to flush those donors with poor responses first (or last) and cancel those with no response. Ultrasonography (Pierson and Ginther, 1988) provides more accurate information about responses than palpation, but currently this expensive equipment can only be justified in research contexts or in large embryo transfer programmes.
Epidural anaesthesia is recommended for non-surgical recovery procedures. The tailhead should be clipped, then scrubbed with iodine soap and swabbed with 70 percent alcohol to prevent infection of the spinal column. The site of the epidural injection is illustrated in Figure 2. A frequent error is to inject too much anaesthetic too far forward, which can cause the cow to lose control of the rear legs and fall down in the chute. We recommend injecting 5 ml of a sterile 2 percent solution of procaine in water using a new 18-gauge needle each time. Good epidural anaesthesia can be monitored by flacidity of the tail.
While the epidural anaesthesia is taking effect, the tail should be secured to one side out of the way, for example, by tying it to a cord looped loosely around the cow's neck. It should not be tied too securely to something stationary, like the chute, for fear of the tail breaking if the cow falls or if personnel forget to loosen the cord before releasing the cow. The rear end of the cow should be cleaned of mud, manure, loose hair, etc., and then the vulvar area scrubbed thoroughly with iodine soap and rinsed carefully with swabs of 70 percent alcohol (Figure 3A). Sufficient time should be allowed for the lips of vulva to dry before inserting the recovery instrument to avoid carrying any alcohol into the uterus; disinfectants are extremely toxic to embryos. For the same reason, an assistant should open the labia gently when the cervical dilator or the recovery device is inserted (Figure 3B).
| FIGURE 2 Site of epidural anaesthesia |  |
Recovery procedures are carried out by manipulation per rectum. Because of the epidural anaesthesia, the rectum can balloon easily due to entry of air during removal and re-insertion of the hand. Once air has entered, it is extremely difficult to work effectively. A simple air pump attached to a length of tubing (Figure 4) to evacuate air from the rectum is an excellent investment, because ballooning of the rectum occurs occasionally, even with skilled personnel. Even so, the best strategy is to prevent entry of air as much as possible.
The basic instrument for non-surgical recovery is the Foley catheter (Figure 5). Generally 18- to 24-gauge sizes are used. It is best to use as large a catheter as can be introduced easily to achieve good rates of flow. Most people prefer two-way catheters, one passage for air and one for fluid, because rates of flow are higher than with three-way catheters, which have two smaller passageways for fluid. The disadvantage of the two-way catheter is the dead space or column of fluid that remains in the catheter during filling and never reaches the uterus. Teflon-coated Foley catheters are recommended to reduce the possibility of embryos sticking to the catheter.
| FIGURE 3 (A) Scrubbing vulvar area with tamed iodine soap; (B) Assistant gently opening vulvar labia to avoid contamination of cervical expander or collection catheter during insertion |  |  |
It is possible to purchase custom-made embryo collection devices. Usually these are very expensive, and cannot be justified. They generally have two advantages over standard Foley catheters in that they are slightly thinner and somewhat longer. For large, older cows, the standard Foley catheter can be lengthened by combining it with a second catheter using glass connecting pieces. Usually this works as well as the expensive, custom-designed catheters.
| FIGURE 4 Length of tubing attached to a vaccum pump to evacuate air from the rectum |  |
| FIGURE 5 Two-way Foley catheters with and without balloon inflated (first appeared in Kuzan and Seidel, 1986) |
 |
Before removal of the Foley catheter from the paper envelope used for sterilization, the air system to the balloon is checked to see that it and the balloon will hold air. The Foley catheter is rinsed with sterile saline, and finally a sterile metal stylet such as the plunger of a Cassou insemination gun (Figure 6) is inserted into the lumen. The system is then ready for insertion into the cow. An assistant parts the labia of the vagina and the device is manipulated through the cervix.
Sometimes great difficulty is encountered in manoeuvring the Foley catheter through the cervix, particularly with heifers of some breeds. As soon as there is a hint of difficulty, the apparatus should be withdrawn and a stainless steel cervical expander (Figure 7) should be inserted first. A gentle, patient technique is essential.
| FIGURE 6 Insertion of stylet to make Foley catheter rigid for inserting through the cervix and positioning in the uterus |
 |
| FIGURE 7 Tip of cervical expander; expander is a 48-cm stainless steel rod, 6.3 mm in diameter, that tapers in the last 4 cm to a 3-mm rounded tip |
 |
There are two fundamentally different approaches to positioning the balloon of the Foley catheter for non-surgical recovery procedures. These are commonly referred to as “body” and “horn” flushes. For the body flush, the catheter is inserted into the uterine body and the balloon is inflated just past the cervix. Some technicians prefer air, others, 0.9 percent sterile NaCl solution to fill the balloon; we recommend air. The single most common error in non-surgical recovery, especially with horn flushes, is overinflation of the balloon. This leads to rupture of the endometrium and loss of flushing fluid (and embryos) into the uterine tissue, from which recovery is impossible. Once this occurs, the only recourse is to reposition the balloon more anteriorly, precluding a body flush. The amount of air used to inflate the balloon usually ranges from 10 to 20 cc, depending on the size of the uterus. The balloon should fit snugly, but should not rupture the endometrium. A disadvantage of the body flush is that the balloon sometimes occludes one of the horns so that only one fills.
For the horn flush, the balloon should be positioned at the palpable bifurcation of the uterine horns (Figure 8). The advantage of the horn flush is that a much smaller volume of the uterus is flushed, which requires less medium and theoretically results in improved embryo recovery rates. This is particularly true for older cows of large breeds with large, pendulous uteri. The major disadvantage of the horn flush is the need to reposition the catheter in the second horn after flushing the first, which requires detaching the inflow-outflow tubing, deflating the balloon and reinserting the stylet. This prolongs the flushing procedure considerably.
In our laboratory, we frequently use a hybrid technique. We start with a horn flush on the side with the largest response and then gently retract the Foley catheter so that the balloon lodges in the uterine body. This sometimes occurs in the course of flushing without a deliberate attempt and illustrates the subtle situation of having enough air in the balloon, but not too much. The second uterine horn is then flushed from the body position by occluding the first horn transrectally or simply allowing both horns to fill and empty.
The uterus can be filled and emptied with either a continuous-flow system or in aliquots, for example, by repeatedly inserting and recovering 50 ml of fluid from a syringe. The same 50 ml may be reflushed two or three times before it is examined for embryos, or each aliquot may be used only once. With the continuous-flow system, the volume of fluid in the uterus is controlled by clamps on the inflow and outflow tubes. The aliquot method is best suited to horn flushes. Both continuous-flow and aliquot methods have staunch advocates.
We recommend the continuous-flow system, which we find less cumbersome (Figures 9 and 10). We place 2 litres of medium in a disposable plastic intravenous infusion bag or Ehrlenmeyer flask held about 1 metre above the cow. This provides the proper pressure from the force of gravity for filling the uterus at the optimum rate. We use 1/4 -inch (inside diameter) Tygon tubing to connect the infusion bag and outflow tube to a Y-connector on the Foley catheter (Figure 10 and Chapter 17).
With either system, the principle is to fill and empty the uterus four to six times. With each successive filling, the uterus tends to expand, particularly in older cows, so that more and more fluid is required for an effective flush.
| FIGURE 8 Position of Foley catheter for uterine horn flush |  |
A good initial reference point is to fill the uterus until the degree of distension is equivalent to a 45-day pregnancy, which takes more fluid with body than with horn flushes and, similarly, more with cows than heifers. With body flushes, horns may be filled alternately or both horns filled and emptied together. Some technicians emphasize massaging the uterine horns to loosen embryos from endometrial folds, while others emphasize expanding the uterus to dislodge the embryos from the folds. Good rates of flow upon emptying are essential for good rates of recovery.
There is also debate about the need to occlude the utero-tubal junction to prevent retrograde flow into the oviduct. This may be a problem if the uterus is fully distended, but otherwise is probably an infrequent occurrence. Nevertheless, we recommend occluding the utero-tubal junction with the thumb and index finger when the filling cycle is at its peak, and gently massaging the uterus with the other three fingers.
| FIGURE 9 Continuous-flow system for recovery of embryos with medium flowing through the Foley catheter into a graduated cylinder. Technician has clamped with a haemostat the inflow tubing from the Ehrlenmeyer flask held above assistant's head | FIGURE 10 Y-connector (A) attached to fluid canal (B) of Foley catheter. Note inflow (C) and outflow (D) tubing and air canal (E) to inflate balloon of Foley catheter |
 |  |
Fluid from the uterus is usually collected in either 2-litre graduated cylinders or through a 75-μ mesh filter (Figure 11). When cylinders are used, the flush fluid is allowed to sediment for 25 minutes. Most normal embryos will settle to the bottom of the cylinder within this time. All but 150 ml is siphoned off with narrow-bore flexible tubing from each cylinder into another cylinder, which is set aside for resiphoning. The bottom 150 ml of fluid is swirled and poured into flat-bottomed dishes scored to divide the bottom into squares (see Figure 12). Each cylinder should then be rinsed at least twice with 20 ml of medium to dislodge any retained embryos.
With the filtration method of isolating embryos, fluid passes through the filter unit and is allowed to escape through a short length of tubing (see Figure 11A); outflow is controlled by means of a clamp. To prevent dehydration, at least 1 cm of medium should be retained in the filter to cover the filter grid on which the embryos rest. To recover the embryos from the filter, one swirls the filter container and pours the contents into a searching dish, and then quickly rinses the filter (Figure 11C) in concentric circles while holding it partly inverted, moving from the outer rim of the grid to the centre, using a 22-guage needle mounted on a 30-cc syringe containing flushing medium without serum or BSA. The omission of protein from the medium for this step is important to prevent foaming when it is ejected from the needle, since embryos are easily lost among the bubbles, which persist for hours. The sides of the filter and the grid should be rinsed several times until all vestiges of mucus and cellular debris are gone. This takes considerable rinsing at high pressure. Medium containing 0.4 percent BSA or 10 percent heat-inactivated serum should be added to the searching dish after the filter has been thoroughly rinsed to keep embryos from floating and sticking to the dish or pipette.
The filter method is considerably faster than the cylinder method, although in the case of about 5 percent of donors, the filter becomes clogged with mucus, so a second filter must be used, and both must be rinsed to recover embryos. With conscientious effort, the cylinder method is just as efficacious as the filter method, and considerably less expensive with respect to materials; however, it is more labour intensive. Thus, it seems to us that the filter method should be used when labour is scarce and expensive and the cylinder method should be used when labour is available and capital is not.
 |  |
| FIGURE 11 Filtration method of isolating embryos from flush fluid illustrating an assembled embryo filter unit (A), the mesh grid on which embryos are retained (B), and rinsing the filter to recover embryos (C) |
 |
The flush fluid should be examined systematically at about 10–14X magnification to locate embryos (see Figure 12). They should be transferred to fresh medium as soon as they are found, and washed through at least three changes of medium (ten changes if there is a chance of export or if infection is suspected; see Chapter 14) as soon as possible. It is a good idea for two people to examine each dish twice. The fluid siphoned from the cylinders should also be resiphoned if that method is used. It pays to be painstaking in searching for embryos. Dishes that are not being examined should be covered and stored where they will not be exposed to excessive light. As soon as all embryos have been located and washed, they should be evaluated and prepared for either immediate transfer or cryopreservation. It is important to record pertinent data promptly to avoid embarrassing errors and loss of information (see Chapter 16).
In some cases reflushing the uterus is appropriate. A second epidural injection is essential. This should be done at least two to three hours after the first flush so that the first epidural block wears off to avoid ballooning of the uterus when equipment is positioned. We have reflushed several hundred donors. Unless clear technical problems have occurred during the first flush, we usually do not recover additional embryos if none was recovered in the first flush. However, we frequently recover a few additional embryos on reflushing if there was a good ovarian response and a good rate of recovery of transferable embryos with the first flush. Reflushing results in about 10 percent more embryos, averaged over many donors.
| FIGURE 12 Examination of flush fluid square by square at 10X magnification to locate embryos. Container is 100×15 mm with 13-mm grid |
 |
Despite good flushing procedures by experienced technicians, not all embryos are recovered. A few are inaccessible in the oviduct, but some are simply missed. The variability in this step is in need of further research. It is particularly frustrating to recover no ova when there is a good ovarian response and no indication of problems with recovery procedures. However, as there were similar situations with surgical recovery of embryos, it is unlikely that the method of recovery is at fault. In our experience, it is best to give donors a luteolytic dose of prostaglandin F2 alpha, or an analogue, after the flushing procedure. Of course, this may not be appropriate for certain infertility cases.
If prostaglandin is not given, donors should be examined ultrasonographically or palpated at 40–50 days after breeding to diagnose potential multiple pregnancy. Even with a luteolytic dose of prostaglandin, an occasional donor remains pregnant. Some recommend two prostaglandin injections at one-week intervals. Note also that most donors return to oestrus later than two to five days after prostaglandin, in contrast to non-superovulated cows.
