Obstetric lacerations are a common complication of vaginal delivery. Approximately 75% of all women who undergo vaginal delivery will experience some degree of perineal laceration.1 Third- and fourth-degree lacerations are those that disrupt the external anal sphincter and are estimated to complicate up to 11% of deliveries.2 Many women with third- or fourth-degree lacerations will develop long-term sequelae such as fecal urgency or incontinence, perineal pain or dyspareunia, or rectovaginal fistulae.3–5 The prevalence of third- and fourth-degree lacerations in the United States has been decreasing during the past 15 to 20 years.6 Prevention of these debilitating complications depends on appropriate recognition of severe lacerations, good surgical technique, and the appropriate choice of suture materials.
With the rise in cesarean delivery rate and decrease in forceps deliveries, resident confidence with and exposure to the repair of obstetric anal sphincter (OAS) lacerations is decreasing.7,8 Formal residency training in OAS repair has been inconsistent, with 60% of obstetrics and gynecology residents in the United States admitting no didactic teaching on episiotomy repair and reporting direct supervision only one third of the time.8
Surgical models have been increasing in popularity and have been shown to improve surgical skills, knowledge base, and confidence.9–17 Several perineal models have been developed to aid in teaching third- and fourth-degree laceration repair to residents. These models range from commercially available anatomic models to cadaveric models to models constructed of common materials such as chicken, sponges, and various animal tongues. Residents have previously reported satisfaction with both the sponge and beef tongue models16,17; however, no study has compared the 2 models. This study aimed to compare the beef tongue and sponge perineal models in terms of improvement in clinical knowledge and repair technique.
Institutional review board approval was obtained before the initiation of the study. Third- and fourth-year obstetrics and gynecology residents at the LAC + USC Medical Center were recruited to participate. All subjects were aware that they were scheduled to participate in a training workshop on OAS injury repair. All subjects took a preintervention written test adapted from Patel et al13 and then attended a 10-minute presentation on evidence-based OAS repair techniques. Subjects were randomized by random number draw to perform the beef tongue model of laceration repair first, whereas the second group was to perform the sponge model of laceration repair first.
The beef tongues were prepared as previously described.11,15 Each beef tongue was cut into 4-in sections (approximately 3 per tongue) (see image, Supplementary Digital Content 1 (SDC 1), http://links.lww.com/SIH/A113, preparation of tongue model). The section of the tongue was incised in the midline to simulate perineal laceration (see image, SDC 2, http://links.lww.com/SIH/A114, tongue model turned on side). A 2-cm triangular section of chicken skin was sewn to the base of the laceration to simulate the rectal mucosa (see image, SDC 3, http://links.lww.com/SIH/A115, preparation of chicken skin; see image, SDC, 4 http://links.lww.com/SIH/A116, section of chicken skin sewn to tongue laceration). The external anal sphincter was simulated using a cut section of encapsulated muscle from the chicken leg, with a central tendon (see image, SDC 5, http://links.lww.com/SIH/A117, dissection of chicken muscle). The tongue was pierced bilaterally with a curved hemostat, and the chicken tendons were grasped and pulled through the tongue until the chicken muscle edges were nearly flush with the “lacerated” tongue tissue (see image, SDC 6, http://links.lww.com/SIH/A118, perforation of tongue; see image, SDC 7, http://links.lww.com/SIH/A119, pulling chicken muscle into location). The chicken skin was incised to simulate the laceration of the rectal mucosa (see image, SDC 8, http://links.lww.com/SIH/A120, completed model). Models were prepared within 24 hours of use and refrigerated. Images of the repaired model are provided for reference (see image, SDC 9, http://links.lww.com/SIH/A121, performance of tongue model; see image, SDC 10, http://links.lww.com/SIH/A122, identification and repair of sphincter; see image, SDC 11 http://links.lww.com/SIH/A123, second-degree laceration repair).
The sponge models were prepared as described by Sparks et al16 Two pieces of felt (pink and white) were glued to the bottom of a colored sponge (see image, SDC 12, http://links.lww.com/SIH/A124, preparation of sponge model glue felt). The sponge was incised to simulate the laceration and a 1-cm2 section of sponge was cut to simulate the external anal sphincter (see image, SDC 13, http://links.lww.com/SIH/A125, incision in sponge to mimic laceration). Images of the repaired model are provided for reference (see image, SDC 14, http://links.lww.com/SIH/A126, performance of sponge model; see image, SDC 15, http://links.lww.com/SIH/A127, sponge model second degree laceration repair; see image, SDC 16, http://links.lww.com/SIH/A128, completed model).
During the performance of the simulated repair, all participants were observed and evaluated using a task-specific checklist (TSC)15 of 14 items and global rating scale (GRS)18–21 (maximum total score, 45) on general surgical skills. The TSC and GRS used are shown in Tables 1 and 2. Participants then performed the alternate model and were again evaluated using the same scales. Participants were asked to rate model realism, ease of use, and comfort level with performing anal sphincter repairs on a 10-point Likert scale. Participants were asked to record overall preference between models on a scale of 1 to 5 where 1 indicates preference for the sponge model and 5 indicates preference for the tongue model. After 3 months, participants were contacted to repeat the written test on anal sphincter repair techniques.
Because this study used a sample of convenience, no power calculation was performed.
All responses were analyzed using a paired or unpaired t test for continuous data, a Mann-Whitney U test for nonnormal or ordinal data, or a Wilcoxon rank sum test for paired nonnormal or ordinal data as appropriate.
A total of 16 residents participated. The CONSORT flow chart is shown in Figure 1. The number of third- and fourth-year residents in each group was equal (50%) in each group. Clinical knowledge scores at baseline were not significantly different between groups (63% vs. 62%, P = 0.77), and previous experience with anal sphincter repairs was similar in each group (median, 2.5 vs. 3; P = 0.56 for the sponge and tongue model first, respectively).
The test of clinical knowledge scores was analyzed using the Shapiro-Wilk test for normality and found to be normally distributed (P > 0.05). All participants performed significantly better on the posttest of clinical knowledge than on the pretest (90% correct vs. 62%, P = <0.001), with no difference in percent improvement between the groups (26% improvement in sponge first vs. 29% improvement for tongue first, P = 0.56). After 3 months, all participants performed significantly better on the posttest of clinical knowledge than on the pretest (79% vs. 62%, P < 0.001), with no difference in percent improvement between the groups (17% improvement in sponge first vs. 16% improvement for tongue first, P = 0.9).
The 2 models were reported to be equally easy to understand (8.4 vs. 8.1, P = 0.38 for beef and sponge model, respectively). The tongue model was felt to be more realistic than the sponge model (8.4 vs. 6.3, P < 0.001). All participants considered the tongue model to be equal or more preferable to the sponge model (median, 4.5; range, 3–5).
There was no difference in the mean score on the TSC for the first model between the groups (sponge first, 93.7% ± 10% vs. tongue first, 92.9% ± 7%; P = 0.84) or for the score on the second model performed between the groups (sponge first, 98.2% ± 3% vs. tongue first, 95.5% ± 7%; P = 0.38). Similarly, there was no change in mean score within each group (sponge first, P = 0.14; tongue first, P = 0.28). There was no difference in the median subscores for knowledge of instruments and suture, knowledge of specific procedure, or overall performance on the GRS for the first or the second model between the 2 groups (Fig. 2).
There was a significant increase in confidence level after performing the first model in both groups (Table 3), with no significant difference between the 2 groups. However, in the group that performed the sponge model first, performing the tongue model second resulted in significant further increase in confidence level (Table 3)
After the first model, only one participant in each group felt no improvement in confidence level. However, after the second model, 4 (50%) participants in the group of tongue model first felt no further improvement in confidence level after completing the sponge model second, whereas all (100%) participants who completed the sponge model first felt significant further improvement in confidence after completing the tongue model second (P = 0.02).
The cost of each model is presented in Table 4. The surgical instruments were obtained at low cost from a local swap meet but are also available through bulk online retailers. It is important to note that these instruments are not surgical quality and care must be taken to prevent them from being turned in for surgical processing. The cost of the sponge model ($3.07) was less than the tongue model ($5.14), and authors’ assessment of the ease of model assembly on a 10-point scale was similar (0 easiest, 10 most difficult; 5.0± 1.4 vs. 6.5 ± 0.7, P = 0.31, sponge vs. tongue model, respectively).
Our study shows that both the sponge model and the beef tongue model are low cost, easy to assemble, and easy to use and result in improvement in clinical knowledge and procedural technique. Residents’ level of confidence improved after performing both models; however, performing the tongue model after the sponge model resulted in further improvement in confidence. This suggests that training with a low-fidelity model followed by a higher-fidelity model may improve resident performance.22 However, performing the sponge model after the beef tongue model did not show the same continued improvement in confidence, suggesting that the additional gain in confidence is likely not due to performing the repair twice but that there may be aspects of performing the beef tongue model that provide a benefit beyond that of performing the sponge model.
Surgical models have been shown to improve resident knowledge and skills of obstetrical laceration repair.10 Several different models have been used, including commercially available anatomic models,9,10 cadaveric pig sphincters,9 beef tongue,13,15,17 porcine tongue,12 and a sponge model.16 The choice of model depends on instructor preference, availability of materials, and cost and local regulations on the use of poultry/animal products within the simulation center. The technical superiority of any individual model is unclear. Only one previous study has compared models, looking at the beef tongue model compared with an instructional video,11 which found that both interventions resulted in an improvement in knowledge on a written examination.
This study is the first to compare 2 surgical models for OAS repair. The strengths of this study include its crossover design and the use of objective outcomes (written knowledge examination, TSC, and objective structured assessments of technical skills). We include information on the costs, ease of use and assembly, as well as the educational utility of both models. Furthermore, this study shows the retention of knowledge 3 months after intervention, although there was no comparison group because all participants had used both models.
Limitations of this study include the small sample size, which may have limited our ability to detect a difference in improvement between the 2 interventions. Furthermore, our written knowledge examination and TSC were not rigorously tested for validity and reliability, although both were based on similar tests published in the literature.13,15 The general rating scale was based on a validated, reliable construct18–20; however, there may have been variation among graders. The variation in grading may have affected the ability of our study to detect improvement between the 2 interventions, and the numbers were too small to accurately assess interobserver reliability. Future directions include application of this model to additional levels of training to see if it has validity across lower and higher levels of training.
Overall, our study shows that both the beef tongue and sponge perineal models are excellent tools to help residents increase their knowledge of and confidence in advanced perineal laceration repair. The use of the models resulted in an increased knowledge immediately after the intervention, which persisted at 3 months of follow-up. Although the models were similar with respect to improvement in clinical knowledge, TSC performance, and objective surgical skills assessment, the tongue model was felt to be more realistic and was preferred by resident physicians. Performing a low-fidelity model followed by a higher-fidelity model saw only incremental benefit. However, if multiple models are to be used, the lower-fidelity model should be performed first. It is clear that performance of both models, coupled with formal didactics, resulted in a durable improvement in knowledge.
The authors are grateful to Nicole Bender, MD, Rebecca Nelken, MD, and Erin Randel, MD, Assistant Professors of the Department of Obstetrics and Gynecology, Keck School of Medicine, University of Southern California for their invaluable assistance in conducting this research study.
1. Elharmeel SM, Chaudhary Y, Tan S, Scheermeyer E, Hanafy A, Van Driel ML. Surgical repair of spontaneous perinal tears that occur during childbirth versus no intervention. Cochrane Database Syst Rev 2011; 8: CD008534.
2. Dudding TC, Vaizez CJ, Kamm MA. Obstetric anal sphincter injury: incidence, risk factors and management. Ann Surg 2008; 247: 224–237.
3. Fornell EU, Mattheisen L, Sjödahl R, Berg G. Obstetric anal sphincter injury ten years after: subjective and objective long term effects. Br J Obstet Gynecol 2005; 112: 312–316.
4. Groutz A, Hasson J, Wengier A, et al. Third- and fourth-degree perineal tears: prevalence and risk factors in the third millennium. Am J Obstet Gynecol 2011; 204: 347.e1–347.e4.
5. Meyer S, Schreyer A, De Grandi P, Hohlfeld P. The effects of birth on urinary continence mechanisms and other pelvic floor characteristics. Obstet Gynecol 1998; 92: 613–618.
6. Handa VL, Danielson BH, Gilbert WM. Obstetric anal sphincter lacerations. Obstet Gynecol 2001; 98 (2): 225–230.
7. Cornet A, Porta O, Pineiro L, Ferriols E, Gich I, Calaf J. Management of obstetric perineal tears. Do obstetrics and gynaecology residents receive adequate training? Results of an anonymous survey. Obstet Gynecol Int 2012; 316983: e1–e7.
8. McLennan MT, Melick CF, Clancy SL, Artal R. Episiotomy and perineal repair: an evaluation of resident education and experience. J Reprod Med 2002; 47: 1025–1030.
9. Andrews V, Thakar R, Sultan A. Outcome of obstetric anal sphincter injuries (OASIS)—role of structured management. Int Urogynecol J 2009; 20: 973–978.
10. Banks E, Pardanani S, King M, Chudnoff S, Damus K, Comerford Freda M. A surgical skills laboratory improves residents’ knowledge and performance of episiotomy repair. Am J Obstet Gynecol 2006; 195: 1463–1467.
11. Davis G. Beef tongue obstetrical laceration model. In: Teaching Model Manual. Atlanta, GA: American Urogynecologic Society Post-graduate Course: Teaching and Evaluating Surgical Skills; 2005.
12. Oyama IA, Aaronoff MC, Burlingame JM. Obstetric anal sphincter injury repair workshop for residents. Hawaii Med J 2009; 68: 133–135.
13. Patel M, LaSala C, Tulikangas P, O’Sullivan D, Steinberg A. Use of a beef tongue model and instructional video for teaching residents fourth-degree laceration repair. Int Urogynecol J 2010; 21: 353–358.
14. Siddighi S, Kleeman S, Baggish M, Rooney C, Pauls R, Karram M. Effects of an educational workshop on performance of fourth-degree perineal laceration repair. Obstet Gynecol 2007; 109: 289–294.
15. Siddiqui NY, Stepp KJ, Lasch SJ, et al. Objective structured assessment of technical skills for repair of fourth-degree perineal lacerations. Am J Obstet Gynecol 2008; 199: 676.e1–676.e6.
16. Sparks RA, Beesley AD, Jones AD. “The sponge perineum:” an innovative method of teaching fourth degree obstetric perineal laceration repair to family medicine residents. Fam Med 2006; 38: 542–544.
17. Uppal S, Harmanli O, Rowland J, Hernandez E, Dandolu V. Resident competency in obstetric anal sphincter laceration repair. Obstet Gynecol 2010; 115: 305–309.
18. Reznick R, Regehr G, MacRae H, Martin J, McCulloch W. Testing technical skill via an innovative “bench station” examination. Am J Surg 1997; 173: 226–230.
19. Winckel C, Reznick R, Cohen R, Taylor B. Reliability and construct validity of a structured technical skills assessment form. Am J Surg 1994; 167: 423–427.
20. Ahmed K, Miskovik D, Darzi A, Athanasiou T, Hanna G. Observational tools for assessment of procedural skills: a systematic review. Am J Surg 2011; 202: 469–480.
21. Martin JA, Regehr G, Reznick R, et al. Objective structured assessment of technical skills (OSATS) for surgical residents. Br J Surg 1997; 84: 273–278.
22. Maran NJ, Glavin RJ. Low- to high-fidelity simulation
—a continuum of medical education. Med Educ 2003; 37 (S1): 22–28.