Semen Extension: A comparison of Methods

By W.E. Morgan Morrow, North Carolina State University - Optimizing procedures for handling semen is an important component of any boar-stud-management program. One procedure that has gained attention is the concept of “pre-extension”, whereby a small portion of extender is added to the collection before the full amount is added.
calendar icon 12 September 2005
clock icon 8 minute read

Dr Morgan Morrow
Swine Veterinary Specialist

Editors Note

At the recent annual meeting of the American Association of Swine Veterinarians in Toronto, Canada, Deborah Murray and three colleagues at the University of Minnesota—Darwin Reicks, D.V.M.; Chris Kuster, D.V.M., M.S., Ph.D.; and Bob Morrison, D.V.M., Ph.D.—compared the efficacy of pre-extended semen to semen that was not pre-extended. Here is their report.


Many boar studs currently pre-extend semen with the thought of reducing cold shock.The most important factors influencing sperm function include collection temperature, storage temperature,and the suspension medium.1 When spermatozoa are cooled too abruptly from body temperature to less than 15 degrees C, cold shock can occur, which may reduce sperm viability.2 Semen should be extended as soon as possible after collection from the boar. Pre-extension is also thought to provide a more gradual osmotic change using a small amount (100 ml) of extender and following up with full extension. Pre-extending semen also quickly exposes the raw ejaculate to antibiotics. Bacteria are a normal component of the boar ejaculate and have been reported to range in concentrations of up to 109 cfu/ml.3

Pre-extension, however is time-consuming in the boar stud. The possibility of full extension in a single step would save time and reduce the potential for error since the semen is handled once instead of two separate times. The aim of this study was to determine if there was a significant difference in reproductive performance between sows that were inseminated with pre-extended semen compared to those that were inseminated with semen that was not pre-extended.

Materials and Methods

Semen was collected from 69 crossbred terminal line boars of various ages housed in the same environment. The ejaculate was collected from each boar using the gloved hand technique into a pre-warmed thermal mug containing a 4-quart plastic collection bag with a built-in gauze filter. The semen was then weighed, and the weight was used to estimate the volume of sperm containing ejaculate (1g=1ml).

The ejaculate was then massaged (to ensure a consistent sample), and a droplet (obtained from a straw) was placed on a pre-warmed glass slide (to 37 ° C) and examined with a phase-contrast microscope to check the motility of the sperm. Motility was graded on a pass or fail basis. If the ejaculate motility failed, the sample was discarded.

One hundred uL of semen were then removed with pipette, added to 2.4 ml sodium citrate, mixed, and the concentration was determined using an IMV Microreader. The microreader sample was checked (25 cells counted) for morphological abnormalities. The total number of sperm was adjusted to disregard sperm with abnormal morphology. Sperm with retained distal cytoplasmic droplets were not included in the adjustment factor.The concentration was then established, and the number of doses and the total extended volume for that ejaculate were determined. The following information (Boar ID, collector, total volume collected, sperm per ml in millions, %coiled tails, %proximal droplets, %motility) was then recorded.

The semen was placed in a warming device which maintained it at a temperature of 37°C. After 2 minutes (from the time the semen was collected), the semen was divided into two equal samples. One sample, labeled E, had 100 ml of extender added to it. The other sample, labeled O, did not receive any extender (Table 1). Both samples were then placed back into the warming device. Six minutes (after collection), both samples were fully extended to the calculated final semen volume for that ejaculate. A drop from each sample (E and O) was checked at 37°C on a glass slide for motility. Motility >70 percent was considered passing, and <70 percent motility was failing. If the ejaculate sample (either E or O) was failing, both samples E and O were discarded. Two to five fully extended semen collections were then combined, according to group (E or O), into a pool. The motility of the pooled sample was checked again using the pass-fail criterion. The pooled groups were then packaged (into pochette bags) and labeled, using a Cryovet automated packaging machine. Semen that was pre-extended received a red label. Semen that was not pre-extended received a blue label. The semen was allowed to cool for 30 to 45 minutes at room temperature and then put into a 16°C room.

Table 1: Extender Groups and Extension Times
Even Group 1: Pre-extension with 100ml at 2 minutes 2: Balance of extender added at 6-8 minutes
Odd Group All of extender added at 6-8 minutes

Once semen was fully cooled, it was eligible to be sent for delivery to the sow farms. Samples of each pool were retained and evaluated at 24, 48, 72, 96, and 120 hours. If semen was determined to be unusable at any time, the sow farms were notified, and the corresponding semen samples from that pool were discarded.

The semen was then distributed to 5 sow farms (totaling approximately 1,700 sows) with equal amounts of O and E semen distributed to each farm. The sow farms followed their standard farm procedure for semen handling, heat- checking, and breeding sows. The semen was used exclusively until it was used up or the farm stopped receiving it. All semen was used within 5 days of the date on which it was collected. Even amounts of O and E semen were used for insemination on each farm by breeding the even-numbered sows with the E-labeled semen and the odd-numbered sows with the O semen. Ninety-eight total pools of semen were used in this trial.

Statistical Analysis

Prior to beginning the study, a prospective power analysis was conducted, which determined that approximately 985 sows would be needed per treatment group to detect a 5 percent difference in farrowing rate and that 645 litters per treatment would be needed to detect a 0.5 pig/litter difference in litter size (power =0.80, alpha =0.05).4 The experimental units were the sows. Independent variables were the treatments (i.e., semen processing technique). Response variables were conception rate, farrowing rate, and litter size (total born). Logistic regression was utilized to investigate if treatment had any influence on conception rate. An analysis of variance (ANOVA) was performed on the litter-size data, with farrowing rate analyzed by Chi-square. Multiple regression and logistic regression techniques were utilized to confirm the consistency of the results by including pertinent sow-level factors in the respective analyses for litter size and farrowing rate (STATISTICA).


There was no significant difference (P=0.12; OR=1.26, 95% CI: 0.94 to 1.70) in conception rate between sows that were inseminated with semen that was not pre-extended (91.8 %) and sows that were inseminated with pre-extended semen (87.7 %). A total of 1,666 sows were included in the farrowing rate analysis, with 1,288 contributing litter size data. No statistically significant differences in either farrowing rate (78.1%, 76.5%; P=0.46) or litter size (11.63, 11.73; P=0.58) were detected between the two methods of semen processing (Table 2).

Table 2: Fertility Results
Treatment Farrowing Rate (SE) N (FR) Total Born (SD) N (TB) Fecundity Index
Pre-extension 76.5% (1.5%) 838 11.73 (3.04) 641 897

The logistic regression analysis confirmed no difference between the treatments for farrowing rate (P=0.27; OR=1.07, 95% CI: 0.83 to 1.05). Likewise, the effect of pre-extension vs. non-pre-extension was not significant (P =0.86) for the multiple regression analysis of litter size (full model R2=0.10, P=<0.0001). However, the variance in litter size was significantly different (P=0.0054) between the two groups (Figure 1).


Farrowing rate and total born were not shown to be different between the pre-extended and non-pre-extended semen. Therefore, no benefit was obtained with pre-extension of semen samples prior to full extension in this study. By not pre-extending semen, costs will decrease with a reduction in worker time spent measuring extender and handling semen. If it takes an additional 30 seconds to pre-extend each semen sample, the time spent daily in a boar stud that collects 100 boars per day is approximately 50 minutes. Weekly and yearly, this totals 5.83 hours and 303 hours, respectively. At a pay rate of $10.00 per hour, the yearly savings are approximately $3,000.

There will also be a reduction of potential error with less overall handling of the semen; such errors include measuring errors, semen spillage, and samples that are possibly pre-extended twice or even miss pre-extension (which would result in an overall inadequate amount of extender added to the semen).

However, the total born of sows inseminated with semen that was not pre-extended was shown to be more variable than total born from sows inseminated with semen that was pre-extended. This variability warrants further study.


1 Johnson, L.A., Weitze,K.F., Fisher,P., and Maxwell,W. M.C. “Storage of boar semen.” Animal Reproduction Science 62:1-3, pp. 143-172. August 2000.

2 Ibid.

3 Althouse, G.C., Kuster, C.E., Clark, S.G., and Weisiger, R.M. “Field investigations of bacterial contaminants and their effects on extended porcine semen.” Theriogenology 53:1167-1176. 2000.

4 Kuster, C.E., Althouse, G.C., and Schaeffer, D.J. “Statistical and biological significance in the use of marketed ‘long-term’ semen extenders” In: Boar Semen Preservation IV, eds. L.A. Johnson and H.D. Guthrie, Allen Press, Inc., Lawrence, Kan., 129-135,2000.

Reproduced Courtesy

Source: North Carolina State University Swine Extension - July 2005
© 2000 - 2022 - Global Ag Media. All Rights Reserved | No part of this site may be reproduced without permission.